I am experiencing a problem with codesigning on Mac OSX (Catalina and older), that I cannot find any answer for. I do have a *.component file, that I am codesigning and notarizing with my Apple Developer ID, which is working perfect. As a post-install step on the target machine, I need to include some installation info within the component's Info.plist (so I am writing back to that *.plist file), which would then qualify as a manipulation of the component itself - ergo, the component's signing is going to be corrupt at this point (the host software will not load my software any more).
So I was trying to follow some advice and do a "blind" codesigning as additional post-install step on the user's machine with:
sudo codesign --force --deep -s - ABC.component
This was just working perfect for a few users, just when I found out, not everyone can make use of this (especially older Mac OSX). Sometimes this command will just not result in a properly signed component. One of the users then found out, he had to update Xcode CL Tools, to sign the file. He went:
xcode.select --install
I have tested this with several other users and it worked with them, too.
Here comes the issue:
As a result, I need to make sure, the user has installed the right version of Xcode Command Line Tools, but a lot of my clients do install on an offline device - so I would not be able to have xcode-select --install as a post-install step (because no internet connection).
Is it somehow possible to bundle the proper Xcode Command Line Tools into my installer and make use of it? Or do you see any other solution for the whole problem?
Related
the problem is already known, that gdb doesn't work with OS X Sierra anymore.
This is really annoying, since I researched already about the solution, and there were some solutions, but they apparently don't work anymore. I tried to use the .gdbinit file solution and the disable csrutil solution, both doesn't work.
Sooo my question is: Does anyone know how to fix it in 10.12.3 ?
Thank you very much!
Looks like there still isn't a solution to this. I don't think "use LLDB instead of GDB" is the ultimate solution. So I went ahead and built a fairly basic docker container which contains the very basics (g++ and gdb).
Here is the Docker Hub container, and the the Dockerfile is on this Github repo.
This solution at least allows you to use GDB on your Mac via a couple simple commands. And since the folder you're coding on is mounted, as soon as you hit save, all your changes appear automatically.
Hope this helps someone else out there.
So thanks to Basile Starynkevitch, who just gave me the idea to update gdb
SOLUTION:
You simply have to update gdb for the latest version (which is 7.12.1 currently). Assuming you have installed Homebrew:
That's what I did:
brew update
brew remove gdb
brew install gdb
It has now updated to the latest version.
Now codesign gdb again with the cert that you used before.
sudo killall taskgated
codesign -fs gdb-certi /usr/local/bin/gdb
launchctl load /System/Library/LaunchDaemons/com.apple.taskgated.plist
Last step is still to use the workaround to create the .gdbinit file:
nano ~/.gdbinit
-> set startup-with-shell off
Debugging works for me now :) (OS X Sierra 10.12.3 - gdb 7.12.1).
UPDATE:
Apparently it will fix certain problems, but not the whole problematic itself. It will still cause errors. Honestly, my solution: When you can: Switch to LLDB and another IDE (Netbeans does not support LLDB). Or use Linux. I currently use Ubuntu with a stick on my MacBook Pro.
None GDB 7.11 or 7.12.1 will not work on Sierra 10.12.4 In short it's because of Apple security upgrade. We need to wait for re-enabling when some new version will shows up.
Easy WA is to use LLDB as a debugger for now. Sample instructions here
Another LLDB instruction here
I found that gdb 7.12.1 did not work for me either. I do not think it includes the crucial commit 82b19a4d2f9c9e8d56f to gdb made on the 9th Nov 2016.
I was able to get the gdb 8.0 development branch to work, however.
Download with, e.g.
git clone --depth 1 --branch gdb-8.0-branch git://sourceware.org/git/binutils-gdb.git
Then go into the checked-out directory
cd binutils-gdb
I have macports so (a) wanted to link to macports libraries and (b) wanted to avoid putting stuff in /usr/local so configured with
./configure prefix=$HOME LDFLAGS="-L/opt/local/lib" CPPFLAGS="-I/opt/local/include"
make -j8
make install
=> puts gdb executable into ~/bin; (static) libraries into ~/lib; documentation into ~/share
Then follow André's tip above to do the code signing; though it will now be
codesign -fs gdb-certi ~/bin/gdb
While helping my friend spin up MinGW and a C++ environment on his Windows 8 computer, I ran the get-mingw script and waited as it ran through all the mirrors for required downloads. However, three downloads completely failed:
libltdl - installer script hung and then moved on after pressing "OK"
automake-1.11 - installer script tried finding 1.10, then 1.9, then 1.8, then 1.7 (all of which failed) until finally settling on 1.6
mktemp - script hung and moved on after pressing "OK"
In all three cases, the script gave me a nice error log upon completion, showing that a majority of packages had been downloaded and installed except for these three, which showed up as errors. However during the installation process I had simply gone to the MinGW sourceforge page and manually found and downloaded each .bin.tar.lzma file that was missing.
Now that I have them, is there a good accepted way to unpack and plug them into my friend's existing MinGW install? In case it's tough, I'm comfortable with unix and dos command line so I'll be able to move executables into the MinGW/bin folder if that's what's needed, I just want to check for the best way to 'fix' the install.
As a side note - even though the error log says these are required packages, adding MinGW/bin/ to the PATH still allows for use of gcc and g++, although not make (possibly because of automake failure?). Is this standard behavior?
Firstly, the package issue can be fixed by using the MinGW installer - keep the packages selected and go to "apply changes" and the script will probably try to redownload the missing packages. I think the original problem was probably just a shoddy wifi connection during repository connection.
However, I then ran into a problem where I tried to run gcc and it gave me a missing -lpthread error ... but this question was able to help me fix that, and gcc and g++ are working fine now (haven't opened and tested Eclipse yet though). Just in case of link decay, the issue I cited arises from the MinGW installer script not downloading the lpthread library upon installation. To fix that issue, quoted from link:
Just run and open MinGW Installation Manager, which should be pre-installed with MinGW, select "All Packages" on the left panel, and on the right panel, search for "mingw32-pthreads-w32" packages and install them.
I think the Installation Manager has libpthread and pthread available for install, and pthread libs were the ones that seemed to solve it for me.
Just recently I have begun to review the code for a project produced by a friend roughly 4 years ago. However, to compile this project I need an outdated version of Qt (version 3.3.8). Much of the code used in this project isn't supported in the same style as Qt4. So I'd first like to obtain Qt3 before I transition to Qt4.
Ubuntu has decided to no longer carry "qt3-dev-tools". Ubuntu has limited the selection to "qt4-dev-tools" in its repository. So installing via the command line option (sudo apt-get install qt3-dev-tools) results in an error:
"Unable to locate package qt3-dev-tools."
So, from here I did some digging for a manual installation? I found a great website to help me through the process, but I'm not very knowledgeable about Ubuntu, or Linux in general. Step 1 was simple. But step 2 and onward stumped me. I'm supposed to be both adding the environment variables listed in step two to the .profile file and then setting them, correct?
In step 2, is logging in again just meant for users with the .login shell? When I type in $HOME/.qt-license of step three, I suppose I'm typing this into the terminal?. Same with the ./configure? Both of these commands return no such file exists.
Does anyone know a better walk through to install Qt3 files, or can give a quick breakdown of this tutorial?
The website tutorial:
http://krm.am.gdynia.pl/doc/qt-3.3.8/INSTALL
If u can get .rpm package then open it in ubuntu software centre . It will get installed.
http://download.qt.io/archive/qt/3/ try downloading the package for qt3 from this qt archive & install it.
I've downloaded dicom3tools in Ubuntu apt-get install dicom3tools, but certain apps are not present.
I've downloaded the source and compiled according to directions on Ubuntu without errors. I have access to most of the apps in the kit, but some just seem to be missing or not compiling.
I need a working binary copy of the pbmtoovl tool from this kit.
Can anyone help me?
Do you know why it is missing?
Do I need to compile differently?
Do you have a copy of the pbmtoovl app pre-compiled?
There is no info on this anywhere on the web, I have nowhere else to turn.
Thanks in advance for any info on this.
Please please help me with this.....
I edited the proper file with a uid.
I ran
imake -I./config -DInstallInTopDir -DUsemyID
and everything looked fine.
make World.
make install
make install.man,
but still no rawtodc or pbmtoovl or any of the dicom creation tools. I really need these tools. Please let me know what I'm doing wrong. On Ubuntu 14 –
I am the author of the dicom3tools debian package. The explanation is given online here.
When you install a debian package, you are required to read the documentation. In this case the documentation was available on your system from:
$ cat /usr/share/doc/dicom3tools/README.Debian
So you'll need to follow the build instructions yourself (see INSTALL):
Edit config/site.p-def to set your UID root (a la UseClunieID, to be
selected with a UseXXXXID define on the imake command line).
NB. Don't ever use any UseClunie*ID or your instances
will conflict with mine !
./Configure
setenv IMAKEINCLUDE -I./config # only needed for suns
imake -I./config -DInstallInTopDir -DUseXXXXID
make World
make install # into ./bin
make install.man # into ./man
I finally did a fresh Ubuntu install, installed xutils, g++, gcc and ran the compiling instructions. It did not install, again, but this time I did have a new directory in bin ending in 'unknown' that miraculously contained all of the compiled binaries. I added that dir to the PATH and VOILA I can access all the tools from the command line....
It's still a problem, but I can now use pbmtoovl
When I try to compile a c++ program in my Mac terminal, I get the following error:
-bash: g++: command not found
but I believe I have all the files I need on my Mac to run the compiler. I have the Developer folder, which contains Xcode in Applications, g++ in the usr/bin folder, and a bunch of other folders.
How can I use these files to install the g++ compiler. Thank you!
Assuming you have the latest Xcode, go to Xcode Preferences, Downloads, and click on the Install button next to Command Line Tools. This should install and put everything into your PATH automatically.
houbysoft's answer above is correct, but the OP needs a little more detail.
The basic idea is that you need the "Command Line Tools" environment installed. But the name for that environment, and how you install it, has changed multiple times over the years. So, houbysoft's steps work for 4.3.3, but not for 3.2.6.
In the 3.2 era, the tools were called "UNIX Development". And, rather than being downloaded and installed from within Xcode, they came as part of the Xcode mpkg and were installed as part of the initial Xcode install. In most versions, there was a checkbox named "UNIX Development", usually checked by default, but in some versions it was a separate step. Looking at the "Xcode 3.2.6 and IOS SDK 4.3" disk image currently available from Apple, it's a checkbox. Anyway, that's the only officially-supported way to get them, but lots of other ways work… Here are your options:
Throw away Xcode 3.2.6 (just trash the whole /Developer) and install 4.3.3, then follow houbysoft's steps to get the Command Line Tools. Unless there's a good reason you can't use 4.3, this is almost certainly the best answer.
Throw away Xcode 3.2.6 and reinstall it, and this time make sure the "UNIX Development" checkbox is on.
Just re-run the 3.2.6 installer, and make sure the "UNIX Development" checkbox is on.
Log into developer.apple.com and look through the downloads for a package named "UNIX Development Tools", "Developer Tools CLI", "CLI Developer Environment", "Command Line Tools", or similar that corresponds to 3.2.6. There is such a package for most, but not all, versions of Xcode, and can be used to set up a command-line build environment either with or without Xcode.
Look at the invisible packages on the Xcode disk image. I don't remember whether you need just DeveloperToolsCLI.pkg, or that plus a few others, but a bit of trial and error or googling should get you there.
If everything is installed correctly, it sounds like all you need to do is get your "PATH" environment variable modified.
This closely related SuperUser question should provide the answers you need, at least if you're running under Snow Leopard. Here's another question about setting environment variables for Lion.