Do these libhdhomerun/hdhomerun_config_gui PKGBUILDs look ok?

I've written a few quick PKGBUILDs for libhdhomerun and hdhomerun_config_gui, which are used to control the HDHomeRun device, a networked ATSC/DVB-T tuner. Mostly just useful for troubleshooting and uploading new firmware; I dunno if I'll submit them to the AUR or not but since they're the first PKGBUILDs I've written I'd like someone a bit more knowledgeable to look over them, since I think they're kind of hackish due to shortcomings in the upstream packages
libhdhomerun:
# Contributor: Garrett Ward <glward1 at gmail dot com>
pkgname=libhdhomerun
pkgver=20081231
pkgrel=1
pkgdesc="Library for interfacing with HDHomeRun device"
arch=('i686' 'x86_64')
url="http://www.silicondust.com/downloads"
license=('LGPL')
source=(http://download.silicondust.com/hdhomerun/libhdhomerun_$pkgver.tgz)
md5sums=('475ee4bd5e0f8ca8dd4259040f6a2f61')
build() {
cd "$srcdir/$pkgname"
make || return 1
install -D -m 755 $srcdir/$pkgname/libhdhomerun.so $pkgdir/usr/lib/libhdhomerun.so
install -D -m 755 $srcdir/$pkgname/hdhomerun_config $pkgdir/usr/bin/hdhomerun_config
hdhomerun_config_gui:
# Contributor: Garrett Ward <glward1 at gmail dot com>
pkgname=hdhomerun_config_gui
pkgver=20081231
pkgrel=1
pkgdesc="GTK2 GUI for configuring the HDHomeRun device"
arch=('i686' 'x86_64')
url="http://www.silicondust.com/downloads"
license=('GPL3')
depends=('gtk2')
conflicts=('libhdhomerun')
source=("http://download.silicondust.com/hdhomerun/hdhomerun_config_gui_$pkgver.tgz" "http://download.silicondust.com/hdhomerun/libhdhomerun_$pkgver.tgz")
md5sums=('5b1ffad1a1c12db1cd27ee3f63898fa9'
'475ee4bd5e0f8ca8dd4259040f6a2f61')
build() {
cd "$srcdir/$pkgname"
./configure --prefix=/usr
make || return 1
install -D -m 755 $srcdir/libhdhomerun/libhdhomerun.so $pkgdir/usr/lib/libhdhomerun.so
install -D -m 755 $srcdir/hdhomerun_config_gui/src/hdhomerun_config_gui $pkgdir/usr/bin/hdhomerun_config_gui
libhdhomerun just includes the library and the old CLI utility, which I was going to just package seperately from the gtk2 gui and add on, but the gtk2 gui doesn't build without building libhdhomerun again; it's hardcoded to look for a ../libhdhomerun directory with the source there and doesn't run if it can't find and (re)build it.  libhdhomerun further lacks an install script altogether, and the one for hdhomerun_config_gui doesn't play nice with fakeroot (installs libhdhomerun.so as a file called lib in /usr and not /usr/lib/libhdhomerun) and further is hardcoded to look in / and will only install to $pkgdir if I configure it with that prefix (--prefix=$pkgdir/usr), which I'm sure will break something.
Anyway, if someone could look over them and tell me how badly I cocked them up, I'd be grateful. Thanks

well, if these package builds are going to be independent of one another, then the second one needs to also grab the source for libhdhomerun.  (you've got two md5sums, but only one source.  that's an error.)
as for the misbehavior of the gtk_gui configuration and build process, is there a TARGETS variable that lets you choose which elements you want to compile?  or --build-with or --enable/--disable switch to turn off rebuilding the library?  alternately, if the build process really does require the library to be rebuilt, but you want the package to only deal with the gui, omit the first 'install' line in the second pkgbuild.
finally, why the weirdness for the build process?  is it written into the ./configure scripts?  the makefiles?  can you patch them?  if so, you can include the patch as one of the sources in the PKGBUILD.
my $0.02.

Similar Messages

  • I have an ipad 2 and I noticed I have over 3000 emails in All Mail. I want to know how can I delete these emails. It looks like you can only do it one at a time. It will take me forever. Please help me.

    I have an ipad 2 and I noticed I have over 3000 emails in All Mail. I want to know how can I delete these emails. It looks like you can only do it one at a time. It will take me forever. Please help me.

    You can only delete them one at a time on the iPad. If you have an IMAP email account, go to your email in on the server using a web browser and select as many as you can at a time and delete them that way. That action will be reflected on the iPad. In IMAP - whatever you do on one device syncs across all devices. If you have POP email - start deleting one at a time.

  • First PKGBUILD - looking for review

    Hello all,
    I have created my first PKGBUILD for some new scientific computing libraries and corresponding examples related to my field. I am looking for input before submitting this to the authors for inclusion on their site (and possibly to the AUR if there ends up being more interest).
    Anyway, here is what I have so far, which seems to work on my system:
    PKGBUILD:
    pkgname=ismrmrd-git
    pkgver=20121030
    pkgrel=1
    pkgdesc="Libraries and examples for the proposed ISMRM Raw Data format"
    url="http://ismrmrd.sourceforge.net"
    arch=('any')
    license=('unknown')
    depends=('hdf5-fortran-cxx'
    'boost-libs'
    'xsd'
    'xerces-c')
    optdepends=('fftw: used to compile example programs'
    'doxygen: only needed if you want to produce your own copy of the documentation')
    makedepends=('unzip'
    'git'
    'cmake')
    # install="${pkgname}.install"
    # source=("http://www.syntax-on.com/static/$pkgname/$pkgname-$pkgver.tar.gz")
    # source=("http://downloads.sourceforge.net/project/ismrmrd/src/${pkgname}_${pkgver}.zip")
    source=("cmakelists.diff")
    noextract=("${pkgname}_${pkgver}.zip")
    md5sums=("9ec7ed899e488961826fa41da7109a84")
    _gitroot="git://git.code.sf.net/p/ismrmrd/code"
    _gitname="ismrmrd"
    build() {
    cd "${srcdir}"
    # unzip ${pkgname}_${pkgver}.zip -d ${pkgname}-${pkgver}
    msg "Connecting to GIT server..."
    if [ -d $_gitname ] ; then
    cd $_gitname && git pull origin
    msg "Local files have been updated"
    else
    git clone $_gitroot "${srcdir}/$_gitname"
    fi
    msg "Received code from GIT or server timed out"
    # stop CMakeLists.txt appending an unwanted dir to the CMAKE_INSTALL_PREFIX
    # also prevent install of /usr/cmake/FindIsmrmrd.cmake
    msg "Patching CMakeLists.txt to comply with Arch Packaging Standards..."
    cd "${srcdir}/${_gitname}"
    patch < ${startdir}/cmakelists.diff --forward
    # start the build
    mkdir "${srcdir}/${_gitname}/build" -p
    cd "${srcdir}/${_gitname}/build"
    msg "Starting cmake..."
    cmake -DCMAKE_INSTALL_PREFIX=/usr ../
    msg "Starting make..."
    make
    package() {
    cd "${srcdir}/${_gitname}/build"
    make DESTDIR="${pkgdir}/" install
    cmakelists.diff
    --- src/ismrmrd-latest/CMakeLists.txt 2012-10-30 11:48:03.332945241 -0500
    +++ src/ismrmrd-latest/CMakeLists_mod.txt 2012-10-30 11:48:27.619612710 -0500
    @@ -10,7 +10,7 @@
    endif (WIN32)
    -set(CMAKE_INSTALL_PREFIX ${CMAKE_INSTALL_PREFIX}/ismrmrd)
    +# set(CMAKE_INSTALL_PREFIX ${CMAKE_INSTALL_PREFIX}/ismrmrd)
    list(APPEND CMAKE_MODULE_PATH ${CMAKE_SOURCE_DIR}/cmake)
    find_package(XSD REQUIRED)
    @@ -69,7 +69,7 @@
    INSTALL(FILES ismrmrd.h ${XSDS_SOURCES} ismrmrd_hdf5.h ismrmrd_hdf5_datatypes.h ismrmrd_export.h DESTINATION include)
    INSTALL(FILES schema/ismrmrd.xsd DESTINATION schema)
    -INSTALL(FILES cmake/FindIsmrmrd.cmake DESTINATION cmake)
    +# INSTALL(FILES cmake/FindIsmrmrd.cmake DESTINATION cmake)
    INSTALL(TARGETS ismrmrd DESTINATION lib)
    INSTALL(TARGETS ismrmrd_xsd DESTINATION lib)
    I'm not using any .install file, is this something big I am missing? Again, this is my first time writing a PKGBUILD, so I appreciate any feedback. Thanks!

    Thanks - I will take note for future packages, or if I end up using a non-git version. makepkg seemed to not be able to handle the zip file itself, though. If I re-hosted the source myself in a tarball it worked, but on the authors' site I could only find it as a zip.
    My main question is still about installation location, though. This typically installs itself in:
    /usr/local/ismrmrd/bin
    /usr/local/ismrmrd/lib
    /usr/local/ismrmrd/include
    /usr/local/ismrmrd/schema
    /usr/local/ismrmrd/cmake
    I was able to change this to:
    /usr/bin
    /usr/lib
    /usr/include
    /usr/schema
    Obviously this prevents cmake/FindIsmrmrd.cmake from being installed at all (not sure this is OK or desirable), It also creates the /usr/schema directory, which I don't think is really standard. Where should I put such files? Should I instead install the whole package into /usr/opt/ismrmrd? It doesn't seem big enough to warrant that, but I don't know where else to put the schema and cmake directories and the associated files. Maybe in /usr/share?
    Last edited by aganders3 (2012-11-07 20:20:40)

  • What are these applications that minimalist looking desktops uses?

    Hi,
    I have started to envy the desktops looking like this: http://ompldr.org/vNmt1dQ , but I can't identify what is being used in order to get a desktop looking like those. Can you tell me what kinds of applications are used in that picture. Like, which window managers, panels, the things in applets, or whatever you can see on that desktop.
    Also, if you are one of the people with those kind of looking desktops, can you tell me which kinds of applications do you prefer in general, along with the ones that are appearant on that picture.
    Thanks in advance.

    yasar11732 wrote:I have found that picture on one of the share your desktop threads, but I am not sure if I could find that one again. And, what I was wondering was more particularly, the bar at bottom that says chat files etc. And the upper bar where volume is shown, and other gauges are also there. What kind of panel is that? And also, the part above the panel list some of the open applications, does that person uses two panels one over each other, or is it something else?
    The gauges and info (top bar and right bottom bar) are probably a combination of programs like conky and dzen2. The left bottom bar is probably inbuilt from the windowmanager, the programs listed above that are probably from tmux or  'screen' (terminal multiplexers that allow multiple 'windows' of terminal programs).
    Last edited by litemotiv (2012-02-12 22:43:33)

  • Do these system temprature's look right??

    I;m wondering if my system is running a little hot.
    Its a MSI-Master2-Far (MS-9620 v1.0) with 2x Opteron 246 2.0Ghz
    with 2x Micron Technology 512mb PC3200 (200 MHz) part number - 18VDDF6472Y-40BG3
    Ive made a screen grab of the CPU-z v1.39 and speedfan v4.32 readouts.
    if people need anymore info please ask.
    Many thanks to you all.

    Check the BIOS i9f it's that bad, just stick in there for like 10 minutes....

  • Motion playback on production monitor through AJA Kona LH card looks nasty.

    Okay, so here's the deal.
    We have recently invested in some shiny new things to equip an edit suite - a Kona LH card and a Production Monitor. Along with a Sony HDV/DVCAM VTR. Specs of our set up are listed at the bottom.
    Pretty much all of the work we have done in the past has been non-braodcast - we're looking to improve workflow and quality of final output. We work mainly with Final Cut Pro, but also with Motion.
    We have connected everything and have marvelled at how much of a difference it makes having a proper video preview from Final Cut (v. 6) as opposed to what we had before - firewire to camera to TV... Of course we were expecting this marked improvement.
    We were also expecting the key benefit of a video preview output from Motion.
    Now. We do get a video output through to the production monitor from Motion. But this is where it gets weird.
    When we do a preview of an animation, it looks really shoddy.
    As a test case, let's take a basic text crawl in Motion 3.
    Static graphics look great, but when you play motion graphics, the image quality is really jaggy and nasty. (even at full frame RAM preview, best render settings, full res - all of these options are apparently properly selected).
    So we did some test exports. Usually we would export animations in the Animation Codec, with an alpha if intended for compositing on top of a video track in FCP... we tried a bunch of different formats - animation, photoJpeg, DVCPRO 50, Aja 2Vuy, Uncompressed 10bit...
    All these play from Quicktime on the computer monitor just fine, but when playing from the AJA TV app to the production monitor, again the same problem. And then when you bring them into Final Cut, the same problem, even when you match the sequence settings and field order to the source video ...
    In effect, it seems that anything generated in Motion just doesn't want to look nice on our Production Monitor - even when imported into FCP.
    HOWEVER - if you drop the Motion PROJECT into an FCP timeline, all is fine!
    We know the AJA card and the Monitor are working because of this fact, and because when you generate titles from within FCP itself - the same text scroll, for instance, it looks perfect!
    I have also found when working with old composited animations originating from AE that you can get FCP to pump these through to the production monitor beautifully when working in an uncompressed sequence setting - or in an Animation sequence setting. (For an Animation sequence to go the the production monitor from FCP, the Video Playback option in A/V Settings needs to be set to "AJA Kona 625i25 RGBA" instead of "AJA Kona 625i25 8 bit" or 10 bit).
    The above point could be of relevance...? The only Video Output options from Motion or AE that apply to our setup and monitor are "625i25 RGBA" and "625i25 DV-RGBA" ... are we supposed to have more options?
    In short, we could work around these issues - it just doesn't seem right however that we can't get a decent external preview from Motion. A big reason for upgrading our setup was faithful video previews of animations from these programs...
    We also can't understand why exports from these animation programs can look so poor. I can export a video with a text crawl generated from a DV PAL Final Cut Project with DV Pal compression settings and it looks massively better playing through AJA TV than our test Motion exports.
    It just isn't right.
    Please help. Pretty please.
    For reference, here are the specs of our set up:
    Apple Dual 2.5 GHz PowerPC G5
    4.5 GB DDR SDRAM
    Graphics Card: ATI Radeon 9800 XT
    Equipped with Kona LH Card
    JVC TM-H1750CG Production Monitor fed via SDI (have also tried component, to no avail)
    It just isn't right.
    Please help. Pretty please. Many thanks in advance.
    For reference, here are the specs of our set up:
    Apple Dual 2.5 GHz PowerPC G5
    4.5 GB DDR SDRAM
    Graphics Card: ATI Radeon 9800 XT
    Equipped with Kona LH Card (version 5)
    2 x 23" Apple Cinema Displays
    JVC TM-H1750CG Production Monitor fed via SDI (have also tried component, to no avail)

    Sounds like you are seeing interlace artifacts - in Motion's project properties, have you tried setting the field dominance to None?

  • What are these features and how can I use them?

    Why do I have SDM Profile, Dyanmic private IP, Pay per message, Mobile Hotspot provisioning,   My mom is on the same account and does not have these.  Everything I look at points to enterprise stuff and administration.  I'm not aware of any of that.  Verizon seems to act like they don't know but it's right on my bill.  My VM seems to be funky too.  I had to change my phone number because of unknown video texts being sent.  Something is still not right.

    If these are items listed on your account as features, I think there are supposed to be there.
    I know I have that dynamic IP address one for my smartphone. So I suspect it is associated with the smartphone provisioning.
    The mobile hotspot provisioning would depend on what kind of plan you have.
    The pay per message is probably associated with your text messages so your phone will receive and send them.
    The other one I'm not sure about, but it could be something associated with the SIM card.

  • Adobe Acrobat X 10.1.13 documents and InDesign CS6 looks very blurry on retina screen

    I received an upgrade to my Adobe creative suite 5.5 to CS6.
    On my macbook pro retina late 2013 15 inch screen, I currently have Adobe CS6 Design Premium, among these apps, I got Adobe acrobat pro, and recently updated my Acrobat Pro to 10.1.13. However, documents in this app have very blurry text.
    I also have adobe InDesingn CS6 version 8.0, and text looks blurry on this as well.
    are there any updates to these apps so they look ok on my retina screen? or does CS6 not support retina?

    Hi missartist,
    Could you please try this out:
    Open Acrobat and go to "Edit -> Preferences -> Page Display"
    Under Smooth Text label, please select 'For Laptop/LCD screens' and click ok.
    Now, check if the text still looks blurry.
    Let me know.
    Regards,
    Anubha

  • NetBeans: Dinamically changing Look and Feel Problem

    Hi there, Java lovers.
    I need some help here!
    I love java, and I used to program my graphical user interfaces by hand. In my applications, there is a dropdown for the user to chose the desired look and feel. Then I save the selected option to a file and on the startup I do something like this:
                UIManager.setLookAndFeel(savedUserSelectedLookAndFeel);That always worked really fine. The look and feel would change and everything worked really well.
    However, programing GUI by hand is really a PITA. So I was seduced by Netbeans. I am using Netbeans 6.0 on Windows XP, and I designed the GUIs really fast now.
    The problem is, the system look and feel works really well, but when I change the look and feel of my application to a different one (Java/Motif/etc..) the look and feel changes but the components positions and sizes becomes all messed up. It really becomes impossible to work with a different look and feel because some components are outside of frames.
    I believe the cause of this is because Netbeans generates the code automatically using Grids and stuff like that, and by doing so, the change of the look and feel becomes messed up. This is a thing that would not happen with my manually created code.
    This is really troubling me. Does someone has the same problem. Is there something I can do? Or do I need to create my GUIs by hand like before?

    Oyashiro-sama (&#23569;&#12375;&#26085;&#26412;&#35486;&#12434;&#30693;&#12387;&#12390;&#12356;&#12427;&#12375;&#12289;&#12381;&#12435;&#12394;&#12300;&#27096;&#12301;&#12387;&#12390;&#12356;&#12358;&#20351;&#12356;&#26041;&#12399;&#12385;&#12423;&#12387;&#12392;&#22793;&#12381;&#12358;&#12391;&#12377;&#12397;&#12288;(^_-) &#12300;&#20887;&#35527;&#12301;&#12387;&#12390;&#12356;&#12358;&#38996;&#25991;&#23383;&#12391;&#12377;&#12290;&#65289;
    Sorry for the delay in reply, but one problem for you is that this is not a NetBeans users forum, thus you will not necessarily receive speedy replies here to your NetBeans questions. The best place to send NetBeans questions is:
    [email protected]
    [email protected] (for NetBeans API issues)
    As for your question, you seem to already know quite a bit about NetBeans UI Look and Feel programming, but have you seen these:
    WORKING WITH SWING LOOK AND FEEL:
    http://java.sun.com/developer/JDCTechTips/2004/tt0309.html#1
    Java GUI Swing Programming Using NetBeans 5 (should apply to NetBeans 6):
    http://wiki.netbeans.org/JavaGUISwingProgrammingUsingNetBeans5
    Regards,
    -Brad Mayer

  • Z_Dialog prog initial input screen-looking like selection screen of Report?

    Hi Experts,
    In internet, I found 2 screens( code is not mentioned there) and mentioned that, they r Dialog Prog. screens, they r as below,
    1 - Initial input screen-programmer mentioned it as a selection screen-1000
    2 - Output screen in ALV grid (Object ALV)
    So, when I saw this initial input screen, it looks like same as selection screen 1000 of a report prog., I mean, its containing all selection screen components, like Select-options with out interval, with out extension, blocks, titles for these blocks!!
    So,pls. let me know that,
    1) How the programmer achived this screen? is he used like,
    SELECTION SREEN 1000 START
    END OF SELECTION SCREEN 1000
    2) How these input fileds r looking like select-optins, parameters,  blocks etc.?
    thanq

    Putting selection screens on dialogue transactions have been covered before in the forum. Please search.
    Rob

  • First PKGBUILD... where/how to install

    I want to create a package for Triangle (http://www.cs.cmu.edu/~quake/triangle.html).
    My current pkgbuild looks like:
    pkgname=triangle
    pkgver=1
    pkgrel=1
    pkgdesc="A Two-Dimensional Quality Mesh Generator and Delaunay Triangulator."
    arch=('i686' 'x86_64')
    url="http://www.cs.cmu.edu/~quake/triangle.html"
    license=('custom')
    provides=('triangle' 'showme')
    conflicts=('triangle' 'showme')
    source=(http://www.netlib.org/voronoi/triangle.zip)
    md5sums=('10aff8d7950f5e0e2fb6dd2e340be2c9')
    build() {
    cd "$srcdir"
    gcc -O -o triangle triangle.c -lm
    gcc -O -o showme showme.c -lX11
    Now, I don't know what to do from here... all the help on the wiki assumes make, make install; whereas this isn't the case here. I think I need to copy the resulting executable to /usr/bin, but I'm not sure at all.
    Any help would be much appreciated.

    fukawi2 wrote:
    anitsua wrote:
    Adding
    install -Dm755 triangle "$pkgdir/usr/bin/triangle"
    install -Dm755 showme "$pkgdir/usr/bin/showme"
    to the build() function works for me.
    To be technically correct, those lines should be added to a package() function
    Shouldn't this function be described in the Creating Packages wiki? There is a "The build() function" chapter but none about the package() function except of it being listed in the example.
    I did miss this function up to now as well.
    BTW: The package() function is missing in the "PKGBUILD Prototype" chapter of the Arch Packaging Standards wiki as well. Is this a new concept?
    Last edited by bernarcher (2010-08-18 07:50:13)

  • My first PKGBUILD

    I want to create a package for iscan, it is a GPL front end written by Epson for their scanners. I used to use it several years ago but it was withdrawn because of licensing issues and the old binaries stopped working on modern linux distros. I have only just discovered that it is now available as source and I want to create a package. This is what I have done so far:
    The PKGBUILD looks like this:
    pkgname=iscan
    pkgver=2.4.0
    pkgrel=1
    pkgdesc="A front end for Epson scanners"
    url="http://www.avasys.jp/english/linix_e/dl_scan.html"
    license="GPL"
    depends=('sane')
    makedepends=()
    provides=()
    conflicts=()
    replaces=()
    backup=()
    install=
    source=("http://lx1.avasys.jp/iscan/2.4.0/$pkgname-$pkgver.tar.gz")
    md5sums=('5fb43d47dacbdae1f19252019a26b6c8')
    build() {
    cd $startdir/src/$pkgname-$pkgver
    ./configure --prefix=/usr
    make || return 1
    make DESTDIR=$startdir/pkg install
    I have run ./configure and make in a temporary directory and everything ran fine.
    I then had a go at making the package and this is what happened:
    [root@avocet iscan]# makepkg
    ==> Making package: iscan 2.4.0-1 (Sat Jan 20 18:33:34 GMT 2007)
    ==> Checking Runtime Dependencies...
    ==> Checking Buildtime Dependencies...
    ==> Retrieving Sources...
    ==> Downloading iscan-2.4.0.tar.gz
    --18:33:34-- http://lx1.avasys.jp/iscan/2.4.0/iscan-2.4.0.tar.gz
    => `iscan-2.4.0.tar.gz'
    Resolving lx1.avasys.jp... 202.41.220.50
    Connecting to lx1.avasys.jp|202.41.220.50|:80... connected.
    HTTP request sent, awaiting response... 404 Not Found
    18:33:45 ERROR 404: Not Found.
    As far as I can make out the actual .tar.gz file is called iscan-2.4.0-0.c2.tar.gz. I understand that hyphens are not allowed in the package name but that is the name of the source tar. Is this what the error is trying to tell me or have I made a different error?
    Thanks.

    Rasi wrote:
    well i compiled it too once.. but now its not working anymore.. ./configure runs fine.. but the make process stops after a few seconds...
    imgstream.cc: In static member function 'static lt_dlhandle_struct* iscan::imgstream::find_dlopen(const char*)':
    imgstream.cc:243: error: 'getenv' was not declared in this scope
    imgstream.cc:247: error: 'getenv' was not declared in this scope
    imgstream.cc:282: error: 'free' was not declared in this scope
    imgstream.cc:287: error: 'free' was not declared in this scope
    make[2]: *** [libimage_stream_la-imgstream.lo] Fehler 1
    make[2]: Leaving directory `/home/carnager/is/src/iscan-2.4.0/lib'
    make[1]: *** [all-recursive] Fehler 1
    make[1]: Leaving directory `/home/carnager/is/src/iscan-2.4.0'
    make: *** [all] Fehler 2
    ==> FEHLER: Build fehlgeschlagen.
    It probably needs some patching. gcc-4.3's c++-Compiler does not include c-headers automatically anymore. Try to find out in what header file 'getenv' and 'free' are defined (man getenv) and include them with
    #include <c headerfilename_without_dot_h>
    For example, if the header file is head.h, include
    #include <chead>
    Maybe you have to to this several times for diverse files and/or headers.

  • My first PKGBUILD - Accentuate 1.0

    Hi everyone, I started to practice some ABS, and i created a simple PKGBUILD to install accentuate (it's a fork of KeyJnote, now unreachable for legal issues), a software to present pdf or png slideshows with interesting (and sometimes hidden) features.
    Because it is the first PKGBUILD I'm going to release in the AUR, It would be great if someone gave me a little review of what I've done.
    Here it is:
    # Contributor: Asa Marco <marcoasa90 <at> gmail.com>
    pkgname="accentuate-git"
    pkgver=20081023
    pkgrel=1
    license="GPL"
    arch=('i686' 'x86_64')
    pkgdesc="Presentation tool based on KeyJNote, with many eyecandy effects"
    depends=('python' 'python-opengl' 'python-pygame' 'pil' 'poppler')
    makedepends=('git')
    optdepends=('mplayer: needed to play audio and videos')
    source=()
    url="http://www.cs.rit.edu/~jrm8005/accentuate.html"
    md5sums=()
    _gitroot="git://repo.or.cz/accentuate.git"
    _gitname="accentuate"
    build() {
    git clone $_gitroot
    cd $startdir/src/$_gitname
    install -D -m755 accentuate ${startdir}/pkg/usr/bin/accentuate
    Another way to make it works, could look like this:
    source=('http://repo.or.cz/w/accentuate.git?a=snapshot;h=heads/master;sf=tgz')
    url="http://www.cs.rit.edu/~jrm8005/accentuate.html"
    md5sums=()
    _gitname="accentuate"
    build() {
    cd $startdir/src/$_gitname
    install -D -m755 accentuate ${startdir}/pkg/usr/bin/accentuate
    Considering that the second option doesn't need any makedepends, what's the best approach?
    Any suggestion is wellcome!

    I like the first way best and having git as an extra dependency isn't really an issue.
    Now, the PKGBUILD looks great (considering it's your first), although there is room for improvement. For starters, you can base your package on /usr/share/pacman/PKGBUILD-git.proto (here's a newer version), which is a nice template for git packages. Doing so will help create a more sophisticated package whilst preventing order inconsistencies. Another thing you might want to do is to remove empty fields from the PKGBUILD (i.e.: source, md5sums, etc), purely for aesthetic reasons. Lastly and more importantly, the license field must be an array instead of a string.
    All three of the above are minor issues and shouldn't put you off.
    Edit: You may as well leave python out of the dependencies list, as it's implicitly required by the rest of the python packages.
    Last edited by foutrelis (2008-10-24 01:53:06)

  • Can make manually but not via PKGBUILD

    I'm trying to install exonerate from the AUR, where it is now an orphan. When I run makepkg -s, it fails.
    The PKGBUILD looked a little bit funny (just old?) so I attempted to expand and modify the build section as follows.
    build() {
    cd "$pkgname-$pkgver"
    ./configure --prefix=/usr
    make
    check() {
    cd "$pkgname-$pkgver"
    make check
    package() {
    cd "$pkgname-$pkgver"
    make DESTDIR="$pkgdir/" install
    This did not help.
    Further, if I download the upstream source files I can run `./configure` then `make` fine, which seeemed to imply to me that the BUILDPKG should work fine. Is there a way to install this via pacman?

    Salkay wrote:
    Scimmia wrote:1. You could, but it's kind of heavy handed. You should mess with the variables and see exactly what's causing the problem. You can then modify the variables before running make.
    I'm having a bit of trouble finding exactly how to do this. Would you have a reference that I could read?
    Also, you say "heavy handed", but if this builds okay, would the resulting compiled file be essentially the same as the files if I tweaked the variables more precisely? What advantage would there be in defining variables more precisely?
    I don't have a reference. You can do it by changing the variables in makepkg.conf, but what I would do is to leave !buildflags in the options, then copy the variable definitions into the build function in the PKGBUILD. The defaults would look like this:
    export CPPFLAGS="-D_FORTIFY_SOURCE=2"
    export CFLAGS="-march=x86-64 -mtune=generic -O2 -pipe -fstack-protector-strong --param=ssp-buffer-size=4"
    export CXXFLAGS="-march=x86-64 -mtune=generic -O2 -pipe -fstack-protector-strong --param=ssp-buffer-size=4"
    export LDFLAGS="-Wl,-O1,--sort-common,--as-needed,-z,relro"
    That should fail to build. You can then comment out each of them to see which one is causing the failure. Once you know that, you can remove the flags one at a time to figure out exactly which one is causing the failure.
    As to what they do, you can check the gcc documentation. Google each of the flags and it should point you to the right part.

  • Received several of these updates and unsure if they're forgeries, FlashPlayer_V.102067514b.exe

    aloha,
    I've received several of the above iterations of presumably flash player updates, however upon checking with the adobe flash url, my V is current.  What are these and who should I be contacting if these are forgeries or malware?
    My OS    WIN7 (x64)
    My Browser     Mozilla Firefox V. 20.0.1 (up to date)
    Adobe flash Player     V.11.7.700.169 (up to date)
    mahalo-n-aloha...konadon

    Aloha Pat,
    Thanks for your interest...
    These have popped up looking like an Adobe flash Player Update splash screen telling me that I need to update.  They occur when I have my browser open and tuned into either Adobe TV, Lynda.com or Creative Cow do some refresher training.  On May 3, I downloaded 6 of these updates.
    I decided to try to load the latest v (the 1 id'ed above) and when I started loading I thought better of it and cancelled the load and began my investigation.
    I have since deleted them from my computer and my browser download history.  the might be in the windows temp file someplace but I don't know if that will give me the URL id they came from.
    I'll try again later to see I if I get them again and try to ID the URL.
    Do you have any further interest?
    mahalo-n-aloha...konadon

Maybe you are looking for