Worked example using 'patch' and 'sed' to make a pkg

I would like this to be reviewed, please. The methododology may need to be standardized a bit more. Also, of course, the way I have used scripts may be a bit controversial. Certainly, anyone who knows what they are doing could easily do without these scripts, but maybe also they are stepping stones to acquiring that sort of proficiency! Also, I would not rule out that the PKGBUILD, itself, can be criticised/ improved with guidance.
Introduction
Here is a method for creating patches for Makefiles which uses sed to convert #PREFIX# to $Startdir/pkg in literal path form, so that the build is fully portable. I asked a while back on the forum whether this was possible. Xentac held my hand, while I picked up the method from him. What follows also incorporates some of his cautions about retaining md5sum integrity. (A point about which I earlier had a weak understanding!) My thanks also to dp who sorted me out on the use of $pkgname and $origname.
Description
The PKGBUILD method is an example of automation, which these scripts follow in that variables are assigned values and then a template-like feel follows on. Despite this resemblance, you will need to make these two files executable by applying
$ chmod +x FILENAME
You will need to execute them in a terminal with the commands
$ cd YOUR_WORKING_DIRECTORY_PATH
$ ./Makefiles_ready
or
$ ./PATCHMAKER
This will all work under fakeroot, provided the files have 'user users' permissions. The attached scripts are annotated examples; and enough is provided so that the reader can fabricate his/her own Makefile_new and test out the method for his(her)self.
I reckon this might be useful practice for learning the technique. I need to mention that for full functionality postfix needs to be installed and set up so that sendmail can work, but a lot of the functionality can be tested without doing this.
The command for starting the application is
$ FaxMail
You can, of course, run it for testing purposes from $startdir/pkg/usr/bin. Note the distinction between $origname and $pkgname, which you will see in the following files.
Getting Directory and Files Ready
You could start by making a new directory in $HOME, possibly called 'TESTBED' and by creating empty files named as per the following, ready for copying and pasting. It could also be useful to create a file called HOWTO and paste these notes into it. You will then need to check the 'user users' permissions and to make the two new scripts exucutable.
Then, you are ready to roll -- completing the whole process with a straight 'makepkg' in the terminal.
File Contents
Makefiles_ready
#! /bin/sh
# Makefiles_ready... generates two clean copies of the Makefile.
origname=FaxMail
pkgver=2.3
### RESULTS:-- two appropriately named copies of clean Makefile.
### 'Makefile_new.scratch' is so named as a safeguard against
### inadvertently overwriting any earlier 'Makefile_new'. So, if you just
### want a clean 'Makefile_pristine' with which to make a new patch,
### delete the generated 'Makefile_new.scratch'.
### REQUIREMENTS:-- Either a downloaded and compressed source file
### in the main directory, and appropriate source array in PKGBUILD.
### Or just the latter.
### NOTES: If there is already a ./src_old, it should first be deleted by
### hand. If there is already a $pkgname-$pkgver.pkg.tar.gz, it will need
### to be removed or deleted. This script will produce a failed make
### but will produce the results we are looking for.
mv ./src ./src_old
makepkg -o
# downloads and extracts files; does not build.
cp ./src/$origname-$pkgver/Makefile Makefile_new.scratch
cp Makefile_new.scratch Makefile_pristine
echo ">> When and if ammended, Makefile_new.scratch "
echo ">> needs to be renamed as Makefile_new. If you do this"
echo ">> and you want to retain your earlier Makefile_new, you "
echo " should first rename it to something like Makefile_prev_new."
PATCHMAKER
#! /bin/sh
# PATCHMAKER... makes a patch file
origname=FaxMail
pkgname=faxmail
pkgver=2.3
### RESULTS:-- $pkgname-$pkgver.patch; md5sums
### REQUIREMENTS:-- Clean Makefile_pristine; Edited Makefile_new;
### PKGBUILD ammended to include $pkgname-$pkgver.patch in source array.
### '#PREFIX#' stands in for '$startdir/pkg' -- see the sed command
### in PKGBUILD which inserts the required literal path, prior to making
### for real.
### It is a good idea to look at the patchfile produced here, and checkout by
### eye that the lines set for deletion are really from the Makefile_pristine.
### The two scripts have worked consistently, according to my tests.
### Makefiles_ready can be used to ensure you have clean makefile copies
### to start from.
diff -aur Makefile_pristine Makefile_new >$pkgname-$pkgver.patch
makepkg -go
## tests sources and produce md5sums; does not start the build process.
echo ">> Check $pkgname-$pkgver.patch"
echo ">> If it is OK, paste md5sums into PKGBUILD"
PKGBUILD
# $Id: PKGBUILD, $
# Maintainer:
origname=FaxMail
pkgname=faxmail
pkgver=2.3
pkgrel=1
pkgdesc="A front end for sending faxes via email."
url="http://wol.ra.phy.cam.ac.uk/FaxMail/"
depends=('tcl' 'tk' 'imagemagick' 'postfix')
makedepends=('tcl' 'tk')
conflicts=()
replaces=()
backup=()
install=
source=(ftp://sunsite.unc.edu/pub/Linux/apps/serialcomm/fax/$origname-$pkgver.tar.gz
./$pkgname-$pkgver.patch)
md5sums=('55dc96aee65ccd3bee2ec38bb6ca0ffd' 'b419add4472576a2a4fc4272e8c7da48')
build() {
cd $startdir/src/$origname-$pkgver
patch Makefile $startdir/src/$pkgname-$pkgver.patch
sed -i "s%#PREFIX#%$startdir/pkg%g" Makefile
mkdir -p $startdir/pkg/usr/bin/ $startdir/pkg/usr/man/man1/
$startdir/pkg/usr/info/
make
make install
faxmail-2.3.patch
--- Makefile_pristine 2004-08-21 20:12:23.000000000 +0100
+++ Makefile_new 2004-08-20 12:28:29.000000000 +0100
@@ -3,19 +3,19 @@
# where can I put the FaxMail wrapper and the tryfax executable
-INSTALLBINPATH = /usr/local/bin
+INSTALLBINPATH = #PREFIX#/usr/bin
# where can I put the FaxMail manual
-INSTALLMANPATH = /usr/local/man
+INSTALLMANPATH = #PREFIX#/usr/man
# where can I put the FaxMail info page. If you don't have gnu info,
# you can set this to /tmp or something. If you do, remember to add
# in an entry to the `dir' file.
-INSTALLINFOPATH = /usr/info
+INSTALLINFOPATH = #PREFIX#/usr/info
# what is the toplevel FaxMail directory.
# Contains the coverage list and FaxMail.tcl script.
-FAXMAIL_DIR = /usr/local/lib/FaxMail
+FAXMAIL_DIR = #PREFIX#/usr/lib/FaxMail
# external program locations
@@ -23,7 +23,7 @@
WISH_CMD = /usr/bin/wish
# where can I find smail or sendmail
-MAIL_CMD = /usr/lib/sendmail
+MAIL_CMD = /usr/sbin/sendmail
# where can I store temporary files
TMP_DIR = /tmp
@@ -39,7 +39,7 @@
CFLAGS =
LDFLAGS =
# Uncomment the following line if linking of tryfax fails.
-#LDFLAGS = -lresolv
+LDFLAGS = -lresolv

I have looked at possibilities for executing
$ makepkg -o -p Makefiles_ready
and
$ makepkg -go -p PATCHMAKER
The second is a serious candidate for doing it that sort of way. The first, however, needs the embedded makepkg and encasing with makepkg therefore executes that process a second time.

Similar Messages

  • Rename files in Terminal using mv and sed

    I want to rename multiple files (in one directory) using mv and sed in Terminal (Mac OS 10.6.8).
    I want to change each and every file name from IMG_nnnn.jpg to HVAC_nnnn.jpg, where nnnn is the camera's sequence number. I want to preserve the sequence number and the jpg extension while changing only the file name prefix (the part before the underscore).
    Here's what I tried:
    macatony-2:HVAC ctossc$ ls -d *.jpg | sed -e 's/.*/mv -f & &/' -e 's/IMG/HVAC/' | sh
    This is the result:
    mv: rename HVAC_0205.jpg to IMG_0205.jpg: No such file or directory
    I wanted "rename IMG_0205.jpg to HVAC_0205.jpg".  The arguments to mv seem to be reversed, so the mv command can't "rename" (or delete) the HVAC* file before it is created.
    Any suggestions?

    I still wonder however, why my original code reversed the order of the mv arguments. Knowing that would make my earlier failed attempts a learning experience.
    1st) nbar's 'for' loop is the best in my opinion
    Why your script got things backwards.  lets break it down
    ls -d *.jpg | sed -e 's/.*/mv -f & &/'
    generates:
    mv -f IMG_1.jpg IMG_1.jpg
    mv -f IMG_2.jpg IMG_2.jpg
    mv -f IMG_3.jpg IMG_3.jpg
    mv -f IMG_4.jpg IMG_4.jpg
    mv -f IMG_5.jpg IMG_5.jpg
    mv -f IMG_6.jpg IMG_6.jpg
    mv -f IMG_7.jpg IMG_7.jpg
    mv -f IMG_8.jpg IMG_8.jpg
    mv -f IMG_9.jpg IMG_9.jpg
    ls -d *.jpg | sed -e 's/.*/mv -f & &/' -e 's/IMG/HVAC/'
    generates:
    mv -f HVAC_1.jpg IMG_1.jpg
    mv -f HVAC_2.jpg IMG_2.jpg
    mv -f HVAC_3.jpg IMG_3.jpg
    mv -f HVAC_4.jpg IMG_4.jpg
    mv -f HVAC_5.jpg IMG_5.jpg
    mv -f HVAC_6.jpg IMG_6.jpg
    mv -f HVAC_7.jpg IMG_7.jpg
    mv -f HVAC_8.jpg IMG_8.jpg
    mv -f HVAC_9.jpg IMG_9.jpg
    because when the -e 's/IMG/HVAC/' executed, it changed the 1st occurance of IMG, which was the IMG following the -f
    I personally do not use 'sed' anymore except for some really trivial stuff.  If I need to general messing with a list or the contents of a file I use 'awk' or 'perl', both of which are a bit easier to work with and for me to read afterwards.  But if you really wanted to fix the 'sed' you could have used:
    ls -d *.jpg | sed -e 's/IMG\(.*\)/mv -f & HVAC\1/' | sh
    HOWEVER, I would not really use either 'awk' nor 'perl' for this task.  I would use something similar to what 'nbar' gave you (although I might have over complicated mine some more where as 'nbar' kept it simple and easy to read
    Message was edited by: BobHarris

  • Looking for working example using javafx.builders.HttpRequestBuilder

    Hi,
    Is there any working example using javafx.builders.HttpRequestBuilder and javafx.io.http.HttpRequest to communicate with application server?
    Thanks in advance.
    LD

    Hi,
    Is there any working example using javafx.builders.HttpRequestBuilder and javafx.io.http.HttpRequest to communicate with application server?
    Thanks in advance.
    LD

  • Photoshop CC. Colour selection not working when using alt and brush.

    Hi, Photoshop CC. Windows 8.1 64 bit. Colour selection not working when using alt and brush. I have tried rebooting and other things like closing swatch panel etc.
    The alt key is otherwise OK.  It's not the recent upgrade from Win 8 to 8.1 (5 days ago) because the colour selection facility was working yesterday. Any ideas.

    I'd say things sound desperate enough to try resetting Preferences
    Reset Preferences
    Windows — Hold down Shift Ctrl Alt immediately after starting Photoshop
    Mac — Hold down Shift Cmd Opt immediately after starting Photoshop
    If that fails, we need to know your operating system?
    Do you have a full CC subscription, or by apps do you mean Photoshop, Bridge and Lightroom?
    Another other information that might pertain ?

  • Have to run windows in safe mode with network to have browserws work correctly. using IE and Firefox. Fist web site comes in correctly next tab just sites wai

    have to run windows in safe mode with network to have browserws work correctly. using IE and Firefox. Fist web site comes in correctly next tab just sites waiting for web site Help

    If it works in Windows Safe mode then you have a problem with other software, possibly security software or a system driver, that is running on your computer.<br />
    It is possible that your security software (firewall, anti-virus) blocks or restricts Firefox or the plugin-container process without informing you, possibly after detecting changes (update) to the Firefox program.
    Remove all rules for Firefox and the plugin-container from the permissions list in the firewall and let your firewall ask again for permission to get full unrestricted access to internet for Firefox and the plugin-container process and the updater process.
    See:
    *https://support.mozilla.org/kb/Server+not+found
    *https://support.mozilla.org/kb/Firewalls
    *https://support.mozilla.org/kb/fix-problems-connecting-websites-after-updating

  • Can someone point me to a small working example using wait()

    Or is wait() what I want to use?
    I need to have my app go to sleep for some period of time, say ten or fifteen minutes, and hide its window during that time. Then after that ten minutes it opens its window back up.
    I've looked all over Google and I can't find any code samples that a newbie could make sense of. Can anyone point me to a sample?
    I wrote what I thought, intuitively, was the way to do it, but I am getting some cryptic message on the "wait()" about "java.lang.IllegalMonitorStateException" which I looked up, but the "explanation" makes even less sense than the raw message. :(
      public void runLoop() {
        getPrefs();
        while (true) {
          ... do stuff ...
          try {
            wait(1000*timeOut);
          } catch ( InterruptedException ie )  {
            ... do stuff ...
      }thanks for any guidance.
    --gary                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                           

    Can someone PLEASE point me to a working example?
    I use Java only very rarely, and only when forced to, and I am not an expert. Frankly, in spite of being expert in a dozen other languages I am COMPLETELY INCOMPETENT in Java. I'm just trying to get this project done and out of my hair so I can safely forget all about Java for the next year or two. I'd give the project to someone else, but every other coder in the shop refuses to go anywhere near Java.
    If I can get this one thing working I can call the project done and be out of here.
    I have a working ap that just needs to close its own window when a button is clicked, for some (user specified) period of time like 10 or 15 minutes, and then reopen its window after that period of time. It is intuitively obvious that I should be able to use timer methods, or sleep() , or wait(), but I have not been able to get any of those to work, and the error messages are more cryptic than ancient Sumerian cuneiform.
    I've Googled for a working example, but haven't found anything that I can figure out. The examples I've found presuppose far more expertise than I posses, and have far too many pieces missing. Can anyone just point me to a complete example of an ap that sleeps for some period of time and then wakes up again?
    I would really, really appreciate the help, and once I've finished this project I promise I won't ever bother you again. Really.
    --gary                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                   

  • Looking for Analog Out example using DMA and C++; cannot find sample "GenMultVoltUpdates_SWTimed"

    Looking for example for Analog Out to multiple channels using DMA and C++.  Found reference to sample "GenMultVoltUpdates_SWTimed" which I do not have and cannot find.

    Hello Gerry,
    When you install DAQmx, there is an option to install
    support for Measurement Studio.  When you
    select this option, it will install these examples onto your hard drive.  All of the examples can be found at the
    following location: (C:\Documents and Settings\All Users\Documents\National
    Instruments\NI-DAQ\Examples\DotNET2.0\Analog Out\Generate Voltage)
    Respectfully,
    Rob F
    Test Engineer
    Condition Measurements
    National Instruments

  • Modify Transition element in work flow using for and not attributes to restrict access

    Hi,
    I have TFS 2013 for my premises and I am working on changing the work flow for TFS work item. I have created a work item as per requirement and it has 3 states
    Active, In Review and Closed in work flow. When developer creates a work item it is in Active state and later it is sent to In Review and then to close after successful review.
    My question is I want development team to be restricted to move the work item to Close state. At present when developer creates a new work item they see Active state. At this point they can not see any other states in the work item. Later they
    fill the details and after checkin they will put that work item to In Review state. At this point they can see Active and In Review states. When they save this work item and modify it again they see Close state as well. I want to design the Transition element
    in work flow in such a way then developers should not be able to see the Close state at all in a work item assigned to them. Only reviewer should be able to transit the work item from In Review to Close.
    I checked the for and not attributes in the transition element but I have no idea about correct syntax I should use to enforce the restriction. I would also like to mention here that I have no separate groups defined in my TFS server. Every
    one is a part of contributor's group. There is no specific group of reviewers and one developer reviews the code of another. I want to design the work flow in such a way that the developer who creates (or gets assigned) the work item should
    not be able to put their own work item in Close state but they should be able to close another developer's work item if that work item has his name in the reviewer's box.
    I have tried to simplify my question by elaborating it hope it is not complex to understand. Let me know if you want further clarification on this.
    Regards, Premal Acharya

    Hi Premal,
    For your scenario, seems the "for" and "not" attributes not are unavailable to make it happens. Because of the developers not in groups, and the same type of work items can be assigned to all the developers hence the attributes cannot
    be allow and deny users in the groups concurrently. Please check the
    page for more information about transition xml element.
    Based on your situation, you can separate the development group and test group. Then you can only allow test group to change the state.
    Best regards,
    We are trying to better understand customer views on social support experience, so your participation in this interview project would be greatly appreciated if you have time. Thanks for helping make community forums a great place.
    Click
    HERE to participate the survey.

  • Using Rescue and Recovery to make backups from 3 different machines

    Hello, I have three Thinkpad laptops, T400, X61 and R61i. I want to back up them using one external HD. The HD is divided into three partitions, the smallest for X61, the medium for R61i and the largest - for T400. I backed up all my laptops into the proper partitions (base backups), but now, then I want  to make partial backups, they all are going to the largest partition (intended for T400). Is it possible (and how) to cause R&R to always use the same partition for each laptop while backing up it?
    Solved!
    Go to Solution.

    Please read the following post:
    http://forums.lenovo.com/t5/ThinkVantage-Technologies/Rescue-and-Recovery-backup-partition-second-ha...
    From this post:
    (1) Learn how to change security/options to peak inside the RRbackups folder
    (2) The conclusion
    outside of an existing RNR folder, For a NEW create RRbackups folder, it will choose the partition with the MOST FREE space.
    ===
    I will offer some commentary and match it to what you are seeing.
    (1) What release of RNR are on the 3 machines? same release? The newer RNR  (been like this for a while), uses the UUID to separate backups within the SAME backup folder.
    (2) for the C drive hdd, you CAN select another  partition on the C drive hdd to hold the rrbackups folder, from the Advanced menu parms. This option is NOT available for second HDD.
    (3) You will always have a RRbackups folder on the C drive.  If you uninstall RNR, the C drive RRbackups is deleted.  ***** Any RRbackups on external hdd will remain***
    (4) What you want to do will not work. RNR is expecting one RNR backups folder on the external hdd, thus implying only one backups folder. From RNR windows, view backups will ONLY show backups that match the UUID of the machine RNR is running on. If you run RNR standalone, you can go to the restore panel and check the "show all backups from all machines" options. This should show all 3 of your machines backups. There is no advantage to have 3 partitiions for 3 backup folder.  Sounds nice, wont work.
    ===
    I will walk you thru a basic how it works on your end.  Correct me if you see something else.
    (1) external HDD
    part 1. small.
    part 2. medium
    part 3. large
    All new partitions. All empty.  Back up large machine, it will chose part 3, create RNR backups folder, Take a FULL base backup. It will store the backup location in the C \RRbackups and the registry. This entry in the folder is named 0.  It should be roughly 1/2 the size of the Large machine C drive. Lets say 39 GB.  Run the next run.  This is an incremental backup (your term = partial).  This entry is named 1, and will be 2 GB., etc.
    You uninstall RNR. Reinstall, change location to Second. Run a backup.  This will be an incremental backup (2) and will continue where it left off. (ie, it should find the old folder with uuid inside).
    (2) the mess
    You backup medium computer. It will find the part3 RNR backups but no matching UUID. humm.. Will it use folder on part3 or choose a largest free space partition, like part 2.    You will have to test.  I would vote for part 3.
    ===
    In summary:
    I believe RNR is designed for only 1 rrbackups folder on an external HDD, regardless of the number of partitions. You seem to indicate you have 3 RRbackups folders on 3 partitions. Look inside each folder.  It might be just using part3 for all your backups and you just dont see it. (ie, were you using part 1 for small machine, took a backup, then added part 2 and part 3, etc). This might work.  Small machine is locked in on small part 1.  Large is locked in on part3. But if you uninstall RNR on small machine, reinstall, take backup.  hum.  It will probably switch over to part 3.  or find part 1.
    This is risk and time you do not need to take.  You want a solid plan that works. Based on this mess, I think RNR folks decided to design only 1 RNR backups folder on external.   

  • I have to constantly re-install flash to get it to work. Using IE9 and Windows 7 64bit

    Hello,
    Thanks in advance for any help.  The last 2 weeks I have had to constantly re-install flash.  I have gone through so many steps as recommended on the forum from using the uninstaller to going through the tool options for enabling and disabling items and to NO END I have to re-install for the day if I want to see any flash websites or youtube.
    Please note that I have IE9 and am on Windows 7 64 bit.  This never use to happen.  Furthermore, I do not want to install google chrome because my outlook did not work well with it.
    I could not open up links and I hope that I do not have to have google chrome to use Flash now?
    Please help it seems that many are experiencing the same?
    Thanks,
    bluicat

    The clean scripts didn't exist at the time, because the installer worked differently. What did exist just the same as today is the caps.db and you need to delete that in C.\Program Files (x86)\Common Files\Adobe\caps.
    Mylenium

  • Joining a work network using wireless and proxy servers

    Hi,
    I have just arrived at a new work location, where we have an afterhours user network that we can connect to in our accommodation. I have had continual trouble trying to connect to the network, and have taken my MacBook to the geeks who provide the network. They can't fix the problem either (and are reluctant to as they don't like Mac's).
    The problem is exactly as follows:
    My airport instantly identifies the network. I need a password to connect to the network name, and this seems to work, but when I run the diagnostics it shows that the airport, airport settings and network settings are all green; but ISP, internet and server are red and failed.
    Next I click on advanced, and the geeks informed me that I need to set up a Web Proxy (HTTP), Secure Web Proxy (HTTPS), FTP Proxy and SOCKS Proxy. They have all been done correctly with the same login and password (which was provided by the geeks). Now they have watched me do this and tried themselves, and they tell me it is correct and has worked previously on other peoples Mac's this exact way.
    But for some reason after applying all this and even restarting the computer just incase, the ISP and onwards still fail to connect.

    The good news is that the basic roaming network setup is the same with the newer 6.x version of the AirPort Utility.
    Here are some step-by-step instructions using the 6.x version of the AirPort Utility.
    First, there are a few key elements to successfully configuring a roaming network, and they are:
    All of the base station must be interconnected by Ethernet. Note: You can use non-Apple routers in this type of network.
    All base stations must have unique Base Station Names.
    All base stations must use the same Radio Mode and Wireless Security Type/Password.
    Each base station should be on a different Radio Channel. Using "Automatic" works well here.
    All base stations, other than the "main" base station, must be reconfigured as a bridge.
    Let's start with the "main" base station. This will be the one directly connected to the Internet modem:
    AirPort Utility > Select the "main" base station > Edit
    Base Station tab > Base Station Name > Enter a unique name here
    Internet tab > Connect Using: DHCP
    Wireless tab > Network Mode: Create a wireless network > Wireless Network Name > Enter the desired name. This will be used on all base stations > Wireless Security: WPA2 Personal (recommended) > Wireless Password > Enter the desired wireless password. This will be used on all base stations.
    Network tab > Router Mode: DHCP and NAT
    Click on Update
    For each additional base station added to the roaming network:
    AirPort Utility > Select the appropriate base station > Edit
    Base Station tab > Base Station Name > Enter a unique name here
    Internet tab > Connect Using: DHCP
    Wireless tab > Network Mode: Create a wireless network > Wireless Network Name > Enter the desired name. This will be used on all base stations > Wireless Security: WPA2 Personal (recommended) > Wireless Password > Enter the desired wireless password. This will be used on all base stations.
    Network tab > Router Mode: Off (Bridge Mode)
    Click on Update

  • Powerpoint 2013 Plugin Example using SDK2 and Gestures

    Hi all
    Does anyone have an example of a powerpoint add-in using VS2013 U4, Powerpoint 2013 and SDK2?
    I am using the Kinect Nuget package but keeping getting an IO exception when loading gesture objects.
    Apparently the Gesture Runtime is having trouble loading AdaBoostTech.dll and RFRProgressTecg.dll even through both are present in the correct bin folder as the Nuget install script sets this up.
    Also tried the older WPF way from the samples using the post build script to copy the dlls 
    xcopy "$(KINECTSDK20_DIR)Redist\VGB\$(Platform)\vgbtechs" "$(TargetDir)\vgbtechs" /S /R /Y /I
    But I still get the DirectoryNotFoundException. Must be something in the way the plugin loads dlls.

    Some additional information:
    I am using firefox 18.0.2.
    I can't copy of the text from the "about:support" page in firefox because when I type that in the address bar, firefox immediately crashes.

  • I have two versions installed on my notebook, 64 bit, win 8.1. Both Versions working, example, psd CC and psd 2014. May I deinstall one version? Thank You, for your answers Sami Tischbein

    Hi guys,
    my notebook runs with both adobe products... The system is win 8.1, 64 bit. The versions cc and the version 2014. May I deinstall one version? But which one? I have no idea, both version are working, but I like to save storage...
    Many Thanks,
    SamiTaschenbier

    SamiTaschenbier you are welcome to remove the previous version if you are no longer utilizing that version.  You can find more details on how to remove Photoshop CC at Install and update apps - https://helpx.adobe.com/creative-cloud/help/install-apps.html.

  • After upgrading to FF 6.0, Yahoo mail looks faded and blotchy and is unusable but it works fine using IE and Chrome.

    A few days ago, I upgraded to FF 6 and Yahoo mail became blotchy and very faded, so much so that it was unusable. I went back to FF 5 and it worked fine. For the heck of it, I just tried FF 6 again and the problem is still there so I'm back to version 5 again. I have a screenshot of what it looked like if that will help.

    Clear Cookies & Cache
    * https://support.mozilla.com/en-US/kb/Template:clearCookiesCache
    Clear the Network Cache
    * https://support.mozilla.com/en-US/kb/How%20to%20clear%20the%20cache#w_clear-the-cache
    Update your Firefox Plugins
    * https://www-trunk.stage.mozilla.com/en-US/plugincheck/
    * http://www.mozilla.com/en-US/plugincheck/
    Check and tell if its working.

  • Using ConnBean and CursorBean with a Data Source

    Hi all,
    I',m making a web app. using the Jdev (RUP4) with OA Extension. I looked in the help menu under "About Data-Access JavaBeans and Tags". Here I found the following example, but for some stange reason I cannot get the setProperty to work.
    Example: Using ConnBean and CursorBean with a Data Source This following is a sample JSP page that uses ConnBean with a data source to open a connection, then uses CursorBean to execute a query.
    <%@ page import="java.sql.*, oracle.jsp.dbutil.*" %>
    <jsp:useBean id="cbean" class="oracle.jsp.dbutil.ConnBean" scope="session">
    <jsp:setProperty name="cbean" property="dataSource"
    value="<%=request.getParameter("datasource")%>"/>
    </jsp:useBean>
    <% try {
    cbean.connect();
    String sql="SELECT ename, sal FROM scott.emp ORDER BY ename";
    CursorBean cb = cbean.getCursorBean (CursorBean.PREP_STMT, sql);
    out.println(cb.getResultAsHTMLTable());
    cb.close();
    cbean.close();
    } catch (SQLException e) {
    out.println("<P>" + "There was an error doing the query:");
    out.println("<PRE>" + e + "</PRE>\n<P>"); }
    %>
    Does anyone know how to set the "Property" to a datasource and make it work?
    Best regards,
    MHCI

    There is no point-and-click (Import Data Source Metadata) way to use an LDAP server as a datasource. You have to use the Java Function provided on dev2dev. If you need help with it, please post here.
    - Mike

Maybe you are looking for