[Solved] Question on how to use (install -d...) in PKGBUILDs

Been looking in on how to use 'install' in PKGBUILDS.  I've looked at the PKGBUILD entries in the wiki and there isn't any mention on this (possibly I missed it??).  I notice that these are in Makefiles and looked on the net but maybe my google-fu is lacking at the moment.  I've got that:
install -d ${pkgdir}/opt
Will create an /opt directory in the pkg directory, and that:
install -m 755 ${srcdir}/${pkgname}-${pkgver}/file ${pkgdir}/opt/
Will install that file in the pkg dir.  So I have a couple questions:
1) How do I install an entire directory into the pkg dir?
2) How do I create a link to /usr/bin?
2) What else is good to know if I have to manually install files/folders???
Here's the current PKGBUILD I'm working on:
# Contributor: Gen2ly <[email protected]>
pkgname=liveusb-creator
pkgver=3.7
pkgrel=1
pkgdesc="Create LiveUSB's from a Linux install image"
arch=('any')
url="https://fedorahosted.org/liveusb-creator/"
license=('GPL2')
makedepends=()
depends=('isomd5sum' 'python' 'pyqt' 'p7zip' 'syslinux')
source=(https://fedorahosted.org/releases/l/i/liveusb-creator/$pkgname-$pkgver.tar.bz2)
build() {
# Just installing in /opt, would /usr/share/python-support be better?
install -d ${pkgdir}/opt
# Install files
install -m 755 ${srcdir}/${pkgname}-${pkgver} ${pkgdir}/opt
Any suggestions?
Last edited by Gen2ly (2009-08-21 09:35:35)

Didn't realize I asked such a complicated question.
I thought that install was a built in command of 'make' that's why I asked but I just found out it has a man page.  It's pretty technical though so I'd wouldn't mind if a someone would help me with basic usage.
Yes, I considered using 'ln' but I've read before that PKGBUILDs should use install instead of cp or mv.  Thinking now that this is not the case with  'ln'???
For installing an entire folder I looked at the install man page that says that -d "treat(s) all arguments as directory names" so would:
install -m 755 ${srcdir}/${pkgname}-${pkgver}/folder ${pkgdir}/opt/
Copy a folder and it's contents?  Newbie question, I know, but first time using install, would appreciate any help.
Last edited by Gen2ly (2009-08-21 07:02:48)

Similar Messages

  • I got new hard driver for my MacBook I don't have the cd but I do have flash drive that has the software I need help because when I turn on my laptop it shows me a file with question mark how can I install the software from the flash driver?

    I got new hard driver for my MacBook I don't have the cd but I do have flash drive that has the software I need help because when I turn on my laptop it shows me a file with question mark how can I install the software from the flash driver?

    Hold down the Option key while you boot your Mac. Then, it should show you a selection of devices. Click your flash drive and it will boot from that.

  • How to use install obj file in tuxedo for win2000

    here i got some compiled tuxedo service files,(*.obj),how can i install
    these service in my tuxedo system on win2000?

    yeah,thanks.:)
    "Peter Holditch" <[email protected]> дÈëÏûÏ¢ÐÂÎÅ
    :[email protected]..
    fish,
    So you compiled a C or C++ service function with
    cl /c [... other options] myservice.c
    and now you have myservice.obj?
    In that case, use
    buildserver -f myservice.obj [ .. other options ] -o myserver.exe
    to generate a Tuxedo server called myserver.exe
    Regards,
    Peter.
    fish wrote:
    here i got some compiled tuxedo service files,(*.obj),how can i install
    these service in my tuxedo system on win2000?

  • How to use installed agent?

    Hi all,
       EM: 12.1.0.2
       Host: RedHat 6.3 64 bits
       DB: 11.2.0.3
       I have installed successfully the Grid Control 12c on HOST01. I suppose that the installation install all the Grid control 12c binaries and an agent in the host where the repository database is. What i want to know is how to use that installed agent to monitor the repository database? I tried to manually add a target to host HOST01, but an error appears to me saying that an agent is already installed on the machine. How to use that agent to monitor my repository database?
    Thanks in advance.

    You can discover (add) the repository database target to start monitoring it using the following steps:
    - Navigate to Setup > Add Target > Add Targets Manually
    - From the Add Targets Manually page, select Add Non-Host Targets Using Guided Process
    - From the Target Types drop-down list, select Oracle Database, Listener and Automatic Storage Management
    - Click Add Using Guided Discovery
    - From the Add Database Instance Target: Specify Host page, select the host where the repository resides, and click Continue
    - Follow the process to discover the repository database
    Regards,
    - Loc

  • In-Design question. How to I install my existing program to my new computer?

    HI,
    I purchased CS6 for the In-Design in Sept '12 and installed it in Oct '12. I have a new computer and it will not read the files. How do I install it to my new computer. I have my WIN Serial Number if that helps.

    You can download all of the CS6 products from this link: Download CS6 products
    Make sure you choose the correct download. If you have a suite, you'll need to download the right one.

  • Question on How to Use LI and NP Data Comparison Report

    Hi,
    We are trying to figure out how to make the Transaction Code PC00_M42_LLPD (LI and NP Data Comparison Report) work. What file type (.xls, .txt, etc) should be used?  Anyone who is familiar with the transaction, please help.
    Best Regards,
    Bry

    Somewhere in the Oracle 8.0 documentation it is stated that one
    enviroment per each thread is required if I want to ensure
    complete concurrency of different threads accessing databases.
    But isn't this a waste of resources since only Handle Alloc &
    Free functions use Enviroment handle, hence they should be
    mutexed for access from concurrent threads.
    But the most time-consuming functions are ServerAttach &
    SessionBegin which shouldn't be mutexed since they have each
    their own Server/Session handle.
    So my question is if a single OCI Enviroment is enough for the
    most demanding tasks or should I create one enviroment per
    thread?
    Most folks are happy with a single environment. The mutex on the
    env handle is only taken when the OCI library is allocating some
    memory for internal operations.
    Tomislav.

  • How to use/install java eclipse

    i've a system that use eclipse as an interface. i try to run it but failed. need to install eclipse first. i've tried but failed. can help me

    previously i tred to install but failed. The
    application i got from one of my friends. he
    developed that program as a cost estimation tool. yes
    it cant pop up after running.Seeing as how you have direct access to the person who developed the program why not ask him why it isn't working ? You would get a far better answer than people here could give - if any, and it will also likely be far faster.

  • Question about How to Use Custom CellEditors

    Hi:
    I have been trying to implement something like this: I have a JPanel, and when you double-click a particular spot on it, I want a textbox to appear. Currently, I am implementing it by using an undecorated JDialog that contains a JTextField in it (such that the text field occupies all the space in the dialog, i.e. you can't even tell that there is a dialog). This seems like a hack to me.
    Another alternative that I have been thinking about is using layered panes, and hiding the textfield below the jpanel, and only bringing it to the front when the user double-clicks.
    Then, I stumbled onto custom celleditors. Now, it seems that everyone is using them in JTables, JLists, JComboBoxes, etc. and I was wondering if it is something that I can use with a JPanel.
    Can anyone comment on the implementations I am considering, as well as using custom celleditors with a JPanel?
    Thanks in advance,
    LordKelvan

    Still don't understand your requirement. Does the text field stay there or disappear after the data is entered? If it disappears, then how do you change the value, do you have to click on the exact same pixel again?
    Maybe you could use a MouseListener and then display a JOptionPane or JDialog to prompt for the information. Or if you want to display a JTextField then add a JTextField to the panel and then when the user enters the data by using the enter key or by clicking elsewhere on the GUI you remove the text field from the panel.

  • Question about: "How to Use Layers in Elements 2.0"

    I have gone over the tutorial "Harness the Power of Layers" and I can not figure out how to open the example the tutorial is showing me. It says: Open the "start.psd" file in the Tutorials folder located inside the Adobe Photoshop Elements Application folder and a wooden doll appears and instructs you how to add the accessories.(hat, bag, shadow, etc.)
    Where do I find or open "start.psd" and the doll with the accessories? Also, is there online instructions that show or animate the steps involved in "layers" using Elements 2.0, which I'd pay for. Thank you, Robert

    Chris, I know this is simple, but I'm having a mental block :-(
    Are you saying go to "My Computer" where I find "C:Program Files/Adobe/Photoshop Element 2/Tutorials ? I can open Adobe Photoshop Elements. I just can't find "start.psd" in the tutorials link... Can you explain it another way ? Thanks
    oh, sorry, yes I'm running Windows XP, 1.5 RAM

  • A question on how to use OCIEnvCreate() function

    In the documentation it is suggested to invoke this function
    before any other OCI call and instead of the OCIInitialize().
    This function also performs any process level initialization if
    required by any mode.
    However, OCITerminate() suggests that it should be called once-
    per-process and as a counterpart of OCIInitialize().
    This is a bit confusing.
    I have a DLL which will be called from different threads and
    provide Oracle data access through OCI. My idea is to allocate
    OCI enviroment only once (when loading the DLL) in multi-
    threaded mode and use it with all the threads/users/sessions.
    Furthermore, I would use OCITerminate() as a cleanup only once -
    when unloading my DLL.
    Can I use those two functions for that? Will they work as I
    expect?

    Somewhere in the Oracle 8.0 documentation it is stated that one
    enviroment per each thread is required if I want to ensure
    complete concurrency of different threads accessing databases.
    But isn't this a waste of resources since only Handle Alloc &
    Free functions use Enviroment handle, hence they should be
    mutexed for access from concurrent threads.
    But the most time-consuming functions are ServerAttach &
    SessionBegin which shouldn't be mutexed since they have each
    their own Server/Session handle.
    So my question is if a single OCI Enviroment is enough for the
    most demanding tasks or should I create one enviroment per
    thread?
    Most folks are happy with a single environment. The mutex on the
    env handle is only taken when the OCI library is allocating some
    memory for internal operations.
    Tomislav.

  • Question on how to use a variable in report object's horizontal formula to shift back and forth at refresh interval.

    Hello All;
    I have an SSRS Report that displays some slider objects. The report auto-refreshes every 5 minutes.
    This report is displayed on a big screen TV and I want to shift the images back and forth every time it refreshes to prevent possible burn-in on the screen.
    I see you can use a formula on the Horizontal position of objects, and I think a variable that gets added 10 and then subtracted by 10 every refresh could be used to display the objects in different positions each iteration, but I don't know how to do that
    in SSRS. I have read up on variables, and I think I need a global variable, but not sure how to update it on each refresh iteration.
    I read the links on using Silverlight to shift things, but that adds a whole other level of complexity.
    I could even do something in the DB and then pull that into the formula field if needed.
    Thanks in advance.
    George Hicks

    Hello All;
    I have an SSRS Report that displays some slider objects. The report auto-refreshes every 5 minutes.
    This report is displayed on a big screen TV and I want to shift the images back and forth every time it refreshes to prevent possible burn-in on the screen.
    I see you can use a formula on the Horizontal position of objects, and I think a variable that gets added 10 and then subtracted by 10 every refresh could be used to display the objects in different positions each iteration, but I don't know how to do that
    in SSRS. I have read up on variables, and I think I need a global variable, but not sure how to update it on each refresh iteration.
    I read the links on using Silverlight to shift things, but that adds a whole other level of complexity.
    I could even do something in the DB and then pull that into the formula field if needed.
    Thanks in advance.
    George Hicks

  • Question on How to Use NHI Difference Report

    Hi,
    We are trying to figure out how to make the Transaction Code PC00_M42_LHID (NHI Difference Report) work. What file type (.xls, .txt, etc) should be used in the Comparison Filemane field?  Anyone who is familiar with the transaction, please help.
    Best Regards,
    Bry

    Hi,
    It is possible that the RFC adapter metadata has not been imported.
    You can verify using the following path in the IR:
    SAP Basis
    --SAP BASIS 7.00
    http://sap.com/xi.XI/System
    Adapter Objects
    Adapter Metadata
    RFC should be present.  If not, ask your basis person to "fix" it.  The adapter metadata can be downloaded fromt the SAP service marketplace.
    Regards,
    Bill

  • [SOLVED] Question about "loadkeys" from the install procedure

    https://wiki.archlinux.org/index.php/In … ard_layout
    https://wiki.archlinux.org/index.php/Be … the_keymap
    # loadkeys ro_win
    They show up as black squares... Even after generating these locales, as per the Beginner's Guide:
    [root@archiso ~]# locale-gen
    Generating locales...
    en_US.UTF-8... done
    ro_RO.UTF-8... done
    ro_RO.ISO-8859-2... done
    Genetation complete.
    What do?
    PS: I noticed that they're activated with "Alt+Shift". There's no mention in the Install Guide or Beginner's Guide about this. Heh.
    Last edited by DSpider (2012-08-24 14:45:34)

    From the LiveCD (in VirtualBox):
    # nano x
    <wrote these characters>
    # cat x | curl -F "sprunge=<-" http://sprunge.us
    And they show up fine: http://sprunge.us/MKBj
    ăîşţâ
    ĂÎŞŢÂ
    Until a better explanation comes along, I've added this info to the BG:
    Don't worry if some of the letters show up as white squares or as other symbols. It's probably just the console font missing those glyphs. The glyph codes are correct and if you press the correct keys they will show up fine when (or if) you set up a graphical environment later.
    https://wiki.archlinux.org/index.php/Be … the_keymap
    I guess I can mark it as solved...

  • Question regarding how to use XHTML format in Form

    I have defined a form where the input interface is iv_text TYPE string and the format of the field in the form is in XHTML.
    But when I set iv_text = '<font face="Times New Roman" color="#ff0000" size="5">Test</font>'. The ADS does not interpret the XHTML and it generates me a PDF file containing this sentence.
    Can someone help me?

    look up for "Using formatters" in the FB help.
    Essentially you'll instantiate a format, NumberFormatter for
    numbers for example, and set it's properties like e.g. showing a
    thousand separator and then one some event of the TextInput --
    valueCommit for example -- you'll do something like:
    ti.text = nf.format(ti.text)
    this is would also give you opportunity to validate user
    input using the same formatter and valid event.
    ATTA

  • How shall i do it ? Complex question about how to use the airport express or time capsule ?

    Hello All,
    Right now i have an access point:
    1/ A mercury router (cheap and old)
    2/ Airport Express
    Do you advice me to use it to replace my cheap old router (mercury) or to pair it with this main old router ? (however i tried to without any success)
    3/ Time Capsule
    Can i use it to relay the signal from the Airport Express to extend the range  of the wifi?
    Can i use it to stream music to my B/W M1 speakers directly ? (I will connect by USB the BW M1 to the time capsule) and would like to be able to play music this way. Is this in anyway possible ?
    Many thanks for your reply !

    ok, i might be wrong but here goes:
    you should attach your modem to your time capsule: your time capsule will act as not only your time machine but also your main wireless transmitter.
    you set it up using "Airport Utility" which is inside the "utilites" folder, which is inside your applications folder.
    this connects your mac wirelessly to the internet.
    to stream music to your speakers you then set up the airport plug to (a) join the existing network (which you set up on the time capsule) and then (b) set it up to stream music to your speakers. again, this is done using "airport utility". do it AFTER you have set up the time capsule.
    you will then be able to use itunes to stream music to your speakers. from itunes on your computer you will be able to chose from the bottom right hand side of the window whether you play the music on your computer or wirelessly into your B+W speakers...
    some notes:
    1. you won't be able to connect the airport plug by USB to your speakers as the airport plug doesn't have a USB output. It has only analogue and optical.
    2. although your time capsule has USB you cannot stream music into it or through it. there's some basic guidance here:
    http://www.apple.com/wifi/
    3. you may find that the sound quality [if you are playing apple lossless files or wav files) isn't as good as if the speakers were connected directly to your mac via USB.  this should be especially noticable because your speakers are so good!
    if the music you play is only mp3 or the stuff purchased from itunes then it shouldn't make too much of a difference.
    4. keep an eye out on the B+W website: there may come a time when you won't even have to use the airport plug, and you will be able to stream music straight to the speakers. i know that B+W have just done that with the new Zepellin speaker that they make.
    hope this helps...

Maybe you are looking for