My machine has lua-5.1.4 installed (from source) but does not have lua.pc. What is the development package to download for lua 5.1.4?
I am new to this myself, so please bear with me while I explain what is 'lua.pc'. If you think some of the information here is incorrect, please make full use of the edit button.
lua.pc or any file with pc for an extension is a file that goes into either /usr/local/lib/pkgconfig/ or /usr/lib/pkgconfig directories. It is a text file containing different fields such as Description etc., information that is used by the make install process. The default installation of lua does not install the lua.pc file. That is because it is the 'normal' package. However, the development package contains this file. The development package creates additional dynamic links (for shared libraries etc.) that are needed during installation of other software where these shared libraries may be dependencies. For more information, read here and here.
Update: lua.pc is present in the etc/ folder within the source tarball.
The lua.pc is in the source tarbal, in the etc folder
Related
I have some task to make linux's bootable image with my own package. This package (named starlet) is a set of .C modules + Makefile. I created the package/starlet directory and added Config.in and starlet.mk; selected in the Buildroot configuration to include my package to build target image.
So, it's works fine...
Now i'm need to build starlet's image with additional library from the zztop-dev package.
zztop-dev package is an .RPM package with set of .H and .C files to build target zztop.a (.so) libraries.
What do I need to do to install zztop-dev.RPM before building STARLET image?
Having the source code for a package stored in a .rpm file is quite uncommon. Buildroot has built-in rules to extract all the most common formats. Using an uncommon format requires you to write extraction rules on your own.
So the first question is whether you can use a more common format that Buildroot has rules for. You probably can access the source code from its original location in a source code repository (git, Subversion, whatever) or a tarball.
If you really need to extract the sources from am .rpm file, then you need to write your own custom extract commands. Look for LIBFOO_EXTRACT_CMDS in the Buildroot user manual.
But if your extract commands call the rpm command to do the extraction then you'll need the rpm tool either installed on your host machine, or packaged as a host package in Buildroot and listed as a dependency of zztop-dev. The former approach is way simpler, but it will force you to have rpm installed on every host machine where you run the build.
In short: This question is basically about telling Linux to load the development version of the .so file for executables in the dev directory and the installed .so file for others.
In long: Imagine a shared library, let's call it libasdf.so. And imagine the following directories:
/home/user/asdf/lib: libasdf.so
/home/user/asdf/test: ... perform_test
/opt/asdf/lib: libasdf.so
/home/user/jkl: ... use_asdf
In other words, you have a development directory for your library (/home/user/asdf) and you have an installed copy of its previous stable version (/opt/asdf) and some other programs using it (/home/user/jkl).
My question is, how can I tell Linux, to load /home/user/asdf/lib/libasdf.so when executing /home/user/asdf/test/perform_test and to load /opt/asdf/lib/libasdf.so when executing /home/user/jkl/use_asdf? Note that, even though I specify the directory by -L during link, Linux uses other methods (for example /ect/ld.so.conf and $LD_LIBRARY_PATH) to find the .so file.
The reason I need such a thing is that, of course the executables in the development directory need to link with the latest version of the library, while the other programs, would want to use the stable version.
Putting ../lib in the library path doesn't seem like a secure idea, not to mention not completely correct since you can't run the test from a different directory.
One solution I thought about is to have perform_test link with libasdf-dev.so and upon install, copy libasdf-dev.so as libasdf.so and have others link with that. This solution has one problem though. Imagine the following additional directory:
/home/user/asdf/tool: ... use_asdf_too
Which gets installed to:
/opt/asdf/bin: use_asdf_too
In my solution, it is unknown what use_asdf_too should be linked against. If linked against libasdf.so, it wouldn't work properly if invoked from the dev directory and if linked against libasdf-dev.so, it wouldn't work properly if invoked from the installed location.
What can I do? How is this managed by other people?
Installed shared objects usually don't just end with ".so". Usually they also include their soname, such as libadsf.so.42.1. The .so file for development is typically a symlink to a fully-versioned filename. The linker will look for the .so file and resolve it to the full filename, and the loader will then load the fully-versioned library instead.
Without packaging system we have (A) source code, which can be translated/compiled to (B) binary code.
In case of debian/ubuntu packages we have (1) source code, (2) source package - dsc file and (3) binary package - deb file. How is it that (2) source package related to (1) and (3)? Why do we need it? And, the most important question: what is the workflow generating (2) and (3) from (1)?
The workflow usually goes approximately like this:
Someone not affiliated with Debian writes some source code and posts it as a package on the web, for example, splint-3.1.2.tar.gz
Someone at Debian downloads the source code, and writes
A set of patch files to make the source build on Debian and conform to Debian guidelines. Run
curl -s 'http://archive.ubuntu.com/ubuntu/pool/universe/s/splint/splint_3.1.2.dfsg1-2.diff.gz' | gunzip -dc | less
to see this for the example package.
A textual metadata file describing the package—this is the .dsc file and debian/control file. “DSC” is an acronym for Debian Source Control.
Binary .deb packages are built for each architecture from the original upstream source code with the Debian-specific patches applied. Here is one such file. The Debian Binary Package Building HOWTO explains the format of these files and how to inspect them.
The .dsc file is not used for build logic, it is more for metadata. However many tools along the way require it. For example, the Build-Depends: field is used to install required build dependencies.
It's actually much more complicated than that. The idea behind Debian packages is that they contain all the information needed to buld a page. Usually, the source is modified to include a debian directory that includes a control file describing the dependencies of that package and other packages that it interacts with (e.g, breaks, replaces, provides virtual package). A rules file explains how to build and install the package. There are also descriptions of how to package since a single source package can become many binary packages (e.g., foo-utils, libfoo0, libfoo-dev). debuild actually reads this information, does the compilation, and produces the binary packages. A subtlety: if foo uses libbar-dev, I may not actually know/care what version of the libbar binary package I use. pbuilder runs debuild in a clean environment so there is no chance of compiling against things you have not explicitly specified.
Consult the Debian New Maintainers' Guide for details.
In the project pandoc, Paths_pandoc is imported in Shared.hs. Paths_pandoc.hs is located in dist/build/autogen/. How is it generated and what does it do for pandoc.
It's a file that is generated by Cabal.
When you specify Data-files: in your .cabal file for your project, those files will be copied to a good location for "data files" on your system when you run cabal install. On Windows, this might be "C:\Program Files\Something" and on Linux it might be "/usr/share/something" (At least when you do a --system install).
Your code needs to know where the files were copied to, so Cabal generates that special module, which contains variables for the install paths that were used to copy the data files, so that your code can find the installed data files.
The module does also contain other information that Cabal provides for you, but the primary purpose of the module is what I just described.
See this blog post for more information.
I am trying to create a Debian package for a Java application.
In my package there is a .jar file which is executable, a script which will run this jar file and a .so file for fmod.
I've read this tutorial.
In the control file there is a 'Depends' field which basically describes the packages that need to be installed in order to install my application. My question is, how do I find which packages are required for my application? I followed the instructions in the tutorial for one of the .so files, and got this:
$ dpkg -S libfmodex64-4.28.09.so
dpkg: *libfmodex64-4.28.09.so* not found.
Also, my application requires Java 1.5 to be installed in order for it to run. How do I specify this in my debian package?
I strongly recommend building your package from source within the Debian packaging infrastructure. Everything will be pretty much automatically taken care of if you use the Ant class in CDBS.
If you do insist on assembling a binary .deb only, equivs is much less hackish than the method described by your document.
You'll want to get the canonical name for your library:
apt-cache search libname
Take care to note the nomenclature at the end of the package. You don't want to specify a specific version in the control file, just the earliest version of the library that is suitable for your application.
You would then use canonical_libname >= major.minor , which lets the system decide if you have (or can update to) the version of the library that can support your application. If you carve this in time, i.e. specifying the full version of your current library, you'll break in the future.
For instance, if you specify libfoo-1.2.34 and future versions of Debian ship libfoo-2.3.45, your package won't install, because it thinks you have an incompatible version of libfoo.