I am creating a Mapnik plugin (https://github.com/rbuch703/coords-mapnik-plugin), and am currently working on packaging it for Debian/Ubuntu. The binary package consists of only a single shared library that is built from C++ code. But being a Mapnik plugin, this library follows conventions quite different from the usual POSIX library conventions:
the file name has to be <name>.input instead of lib<name>.so
the file is installed in the Mapnik plugin directory (usually /usr/lib/mapnik/input)
the file is not supposed to be found by ldconfig, but rather Mapnik tries to find the plugin by itself at runtime
Now the plugin's build system is cmake, which makes most parts of Debian packaging straight-forward: the debian/rules file contains only the basic lines:
#!/usr/bin/make -f
%:
dh $#
However, I am running into problems with the substitution variable {shlibs:Depends}: it is simply not set (in particular, there is no corresponding line in the debian/<package name>.substvars file), and Lintian rightly complains about that fact (Lintian's actual complaint is missing-dependency-on-libc. But when I manually add a libc dependency, Lintian explains package-depends-on-hardcoded-libc, which means "The given package declares a dependency on libc directly instead of using ${shlibs:Depends} in its debian/control stanza."). I would like to satisfy Lintian in than respect, but are unable to do so.
Now I found that I could add the line
dpkg-shlibdeps debian/<packagename>/usr/lib/mapnik/input/coords.input
to my rules file. That will create the correct ${shlibs:Depends} line, but it will create it in the wrong file (debian/substvars instead of debian/<package name>.substvars), where the build system simply ignores it and Lintian keeps complaining about missing dependencies.
I am guessing that the root of my problem is that my Mapnik plugin does not conform to the POSIX library naming conventions (and as a Mapnik plugin cannot do so), and thus the packaging system does not handle it correctly. But I am at a loss as to how to fix this problem.
Additional notes:
the packages are built using debuild. Apart from the Lintian error messages, the build process work fine and correctly creates the .deb package.
my practical goal is for the package to build cleanly on Launchpad, so that I can add it to my Ubuntu PPA.
you can provide an output file for dpkg-shlibdeps with the -T flag.
something like:
override_dh_shlibdeps:
dh_shlibdeps
dpkg-shlibdeps \
-Tdebian/<packagename>.substvars \
debian/<packagename>/usr/lib/mapnik/input/coords.input
if there are multiple *.input files, you could also do something like:
override_dh_shlibdeps:
dh_shlibdeps
find debian/<packagename>/ -name "*.input" -exec \
dpkg-shlibdeps -Tdebian/<packagename>.substvars {} +
Related
This is very close to what I'm trying to accomplish. How to compile and run kotlin program in command line with external java library
I really want to learn how to compile and run simple code that includes libraries but am getting a bit lost when it comes to including classpaths.
I’m currently trying to compile and run
import kotlinx.serialization.*
import kotlinx.serialization.json.*
#Serializable
data class Project(val name: String, val language: String)
fun main() {
// Serializing objects
val data = Project("kotlinx.serialization", "Kotlin")
val string = Json.encodeToString(data)
println(string) // {"name":"kotlinx.serialization","language":"Kotlin"}
// Deserializing back into objects
val obj = Json.decodeFromString<Project>(string)
println(obj) // Project(name=kotlinx.serialization, language=Kotlin)
}
using
kotlinc -cp "C:\PROGRA~1\Kotlin\lib\kotlinx-serialization-runtime-1.0-M1-1.4.0-rc.jar" main.kt
to compile with this compiler
https://blog.jetbrains.com/kotlin/2020/07/kotlin-1-4-rc-released/
allowed lib at bottom of the article
that's where kotlinx-serialization-runtime-1.0-M1-1.4.0-rc.jar is coming from. I chose this runtime jar because when I use the new kotlin 4.0.21 compiler it requires the kotlin-serialization-runtime-1.0.1.jar which you need to build yourself but when I download the source and run gradle build it doesn't seem to get generated (separate problem but would love to know how to build the runtime jar myself)
when I try and run I get
Exception in thread "main" java.lang.NoClassDefFoundError: kotlinx/serialization/json/Json
at MainKt.main(main.kt:12)
at MainKt.main(main.kt)
at java.base/jdk.internal.reflect.NativeMethodAccessorImpl.invoke0(Native Method)
at java.base/jdk.internal.reflect.NativeMethodAccessorImpl.invoke(NativeMethodAccessorImpl.java:64)
at java.base/jdk.internal.reflect.DelegatingMethodAccessorImpl.invoke(DelegatingMethodAccessorImpl.java:43)
at java.base/java.lang.reflect.Method.invoke(Method.java:564)
at org.jetbrains.kotlin.runner.AbstractRunner.run(runners.kt:64)
at org.jetbrains.kotlin.runner.Main.run(Main.kt:149)
at org.jetbrains.kotlin.runner.Main.main(Main.kt:159)
Caused by: java.lang.ClassNotFoundException: kotlinx.serialization.json.Json
at java.base/java.net.URLClassLoader.findClass(URLClassLoader.java:435)
at java.base/java.lang.ClassLoader.loadClass(ClassLoader.java:589)
at java.base/java.lang.ClassLoader.loadClass(ClassLoader.java:522)
... 9 more
I know I need to include a classpath when I run
kotlin MainKt
but have tried everything with no success
I've tried many different combinations of things including
compiling with and without
-Xplugin="C:\PROGRA~1\Kotlin\lib\kotlinx-serialization-compiler-plugin.jar doesn't seem to make a difference so I left it off.
I have tried compiling to both a java .jar as well as a kotlin .class file both seem to need classpath information at runtime. I would rather compile to a kotlin .class and keep java out of this until I really need it. This way I can learn what java is really doing in my application.
I guess what I really want to know is how one can determine what is required at runtime for an executable to run. I found this site which helps show dependencies but is for older versions of kotlin https://kotlin.binarydoc.org/org.jetbrains.kotlin/kotlin-compiler-dist/1.3.71/package?package=kotlinx.serialization
I’ve also been peaking into the .class files using https://github.com/google/android-classyshark
and
https://github.com/borisf/classyshark-bytecode-viewer
but still when people tell others, on StackOverflow, what classpath they need to use to solve their problem it seems like magic. Can someone out there teach me how to fish without gradle?
p.s. If anyone has any good resources on learning the internals of how gradle is building the project. I've looked here a bit https://docs.gradle.org/current/userguide/userguide.pdf but didn’t seem to help. maybe I missed something. Also, this page https://kotlinlang.org/docs/reference/serialization.html#example-json-serialization seems to have what I need but can't seem to transfer that to what the command line needs.
dependencies {
implementation("org.jetbrains.kotlinx:kotlinx-serialization-json:1.0.1")
}
what does this mean? I think this is referring to this https://github.com/Kotlin/kotlinx.serialization
but then how can I build what I need from this repo and use it to allow my application to know where its runtime dependencies are? ugh. I really wanted to figure all this out myself, but I must kneel to the internet gods for this one. Sorry, my post is a mess. I love to learn.
It took some time but I was able to build and run the serialization sample found at https://github.com/Kotlin/kotlinx.serialization on the command line using the current kotlinc compiler and the kotlinx.serializtion.1.0.1 library.
Here are the direct links to the compilers and libs
kotlinc and kotlinc-native v1.4.20
https://github.com/JetBrains/kotlin/releases/tag/v1.4.20
Kotlinx.serialization v1.0.1
https://github.com/Kotlin/kotlinx.serialization/releases/tag/v1.0.1
These both can also be found in the 1.4.20 releases blog post under the section titled How To Update:
https://blog.jetbrains.com/kotlin/2020/11/kotlin-1-4-20-released/
Setting Up Katlin’s .jar Libraries
After updating my path to point to the new compilers I still needed to build the serialization libs. This was as simple as running gradle build in the root directory of the unzipped kotlinx-serialization-1.0.1 folder. Make sure to set your JAVA-HOME system variable before you do this or it won’t work.
Once it's built you need to grab both the kotlinx-serialization-json-jvm-SNAPSHOT-1.0.1.jar and the kotlinx-serialization-core-jvm-SNAPSHOT-1.0.1.jar files and move them into the project directory. This definitely confused me because I had found a runtime lib for kotlinx serialization on the MVN repository site that was one jar file, but I wasn't seeing it after building the 1.0.1 libraries. Once I extracted the 1.0.1 runtime jar I found online, by renaming the .jar to .zip, it became apparent that it consisted of both the contents of the core and json jars. Don’t use the kotlinx-serialization-1.0.1-SNAPSHOT.jar. This jar only contains a blank MANIFEST.ms file. You can find the kotlinx-serialization-core-jvm-1.0.1-SNAPSHOT.jar in the kotlinx.serialization-1.0.1\core\build\libs folder and the kotlinx-serialization-json-jvm-1.0.1-SNAPSHOT.jar in the kotlinx.serialization-1.0.1\formats\json\build\libs folder. anyways.
Compiling Your .jar Library
once you have the jars in your project folder you can build your project
I included my cleanbuildandrun.sh shell script down below for easy reference.
My first attempt 1) was to try and build the project without compiling it to a .jar library file. This was a complete failure. I got it to compile but running the project proved much harder. I was unable to tell kotlin where the libraries were at runtime. I tried so many different things Including trying to point it to a manifest file I created but nothing seemed to work. It seems you need to build an executable jar in order to make this work. which brings me to my second try 2). This is where I found more success.
Attempt 2)
First you need to include the kotlinx-serialization-compiler-plugin.jar using the "-Xplugin" compiler flag. My understanding is that plugins are used to define annotations to the compiler like #Serializable. You can find this jar file in the lib folder inside the compiler you just downloaded. I copied this into my projects /lib folder next to the other jar files to make things self-contained and portable.
Next you need to tell the compiler where to find the library classes you want to access using the "-classpath" or "-cp" compiler flag.
Make sure to include kotlin runtime libraries using the "-include-runtime" compiler flag. This will bundle the kotlin standard class libraries within your jar so you don’t need to point at them during runtime.
Last direct the compiler to build a jar file by providing the -d compiler flag with the name and extension of your soon to be .jar file. That’s it, your off compiling.
Example Shell Script:
#!/bin/bash
sh clean.sh
case $1 in
1) # Comming Soon
kotlinc -verbose -Xplugin="lib\kotlinx-serialization-compiler-plugin.jar" \
-cp "lib\kotlinx-serialization-json-jvm-1.0.1.jar;lib\kotlinx-serialization-core-jvm-1.0.1.jar" \
main.kt
;;
2) # Working
kotlinc main.kt -Xplugin="lib\kotlinx-serialization-compiler-plugin.jar" \
-cp "lib\kotlinx-serialization-json-jvm-1.0.1.jar;lib\kotlinx-serialization-core-jvm-1.0.1.jar" \
-include-runtime -d main.jar
jar ufm Main.jar ManifestAdditions.txt lib
kotlin main.jar
;;
3) # Comming Soon
kotlinc-native main.kt -verbose -Xplugin="lib\kotlinx-serialization-compiler-plugin.jar" \
-cp "lib\kotlinx-serialization-json-jvm-1.0.1.jar;lib\kotlinx-serialization-core-jvm-1.0.1.jar" \
-manifest ManifestAddition.txt -o main
;;
esac
Running your .jar Library
By default, when you compiled the jar it created a MANIFEST.ms file that it uses to tell your jar library where the entry point is. which would be enough if we weren’t using additional libraries in our application. So next we need to add those libraries to the jar file we compiled while at the same time updating its MANIFEST.ms file to tell it where those libraries are within that jar file. We can use the cli tool jar to accomplish this. With the command:
jar ufm Main.jar ManifestAdditions.txt lib
we are able to update the current jar file.
u - tells jar we want to update an existing jar file
f - indicates that we are providing the jar file we want to update on the cmd line
m - indicates that we will be providing the manifest file
The Manifest .txt file should look like this:
Main-Class: MainKt
Class-Path: lib\kotlinx-serialization-core-jvm-1.0.1.jar lib\kotlinx-serialization-json-jvm-1.0.1.jar
Make sure to add a new line at the end of the file or it won’t parse the Class-Path.
That’s it. Now we have an executable jar file that we can use to run our serialization code on the command line:
kotlin main.jar
should output:
{"name":"kotlinx.serialization","language":"Kotlin"}
Project(name=kotlinx.serialization, language=Kotlin)
Post Mark
I would really like to turn this answer into a blog post that explains how to use the kotlin compiler, with libraries, on the command line. The information is out there but it seems to be scattered. I would like to include how to compile and run without using jar files, if that’s even possible, as well as how to compile and run using the native compiler. If anyone can help fill in these gaps it would be much appreciated. I think this information could help others learn how to set up simple test environments so they can better understand the functionality of these libs without having to set up a build script. This will be my first attempt at creating a tutorial type blog post so any information would really help.
As a package manager for a Linux distribution, I want to install docs into a separate prefix. With CMake projects, the docs installation location is controlled by CMAKE_INSTALL_DOCDIR from GNUInstallDirs module. Unfortunately, unlike the other directory variables, this one contains the project name so I cannot just use cmake "-DCMAKE_INSTALL_DOCDIR=$myDocPrefix/doc".
With GNU Make, I would run make "DOCDIR=$myDocPrefix/doc/\$(PROJECT_NAME)" and have Make interpolate it but the documentation of CMake’s -D option does not mention interpolation and I understand that CMake uses much more complex system of cache entries where interpolation might be problematic (especially if the referenced variable is not yet in cache).
I could pass tailor-made CMAKE_INSTALL_DOCDIR to each CMake project but would be bothersome as I would have to do that in every package definition manually; being able to define configureCmakeProject function and have it take care of everything automatically would be better. When setting it manually, I would also want to make sure it matches the PROJECT_NAME of the respective CMake project – well, I could resign on that and just use $packageName from the package definition instead but keeping packages as close to upstream as possible is preferred.
Alternately, I could try to grep CMakeLists.txt for project command but that seems fragile and might still result in misalignments. I doubt it is possible to extract it using some CMake API since the project is not configured at the time and we actually need the value to configure the project.
Is there a way I can configure CMAKE_INSTALL_DOCDIR to use custom prefix but still keep the project name set by the CMake project?
I have a small CMake project with different Build Types debug and release. I'm also providing a Debian package for this project. Building the Debian Package for release and providing it on my own Debian repository works perfect.
Now I also want to provide another Debian package for debug, due to debugging purposes, with a different package name. For example, my project is called myproject, and the debugging package should be myproject-debug.
I already read documentation about how to solve this in the debian/control file. I want to use Replaces: ... on each package vice versa, so that you can install only one of the both packages at a time. So either myproject or myproject-debug, but not both at the same time, to use the exact same files and filenames but only the binary has more debugging informations and debug prints in the myproject-debug package. Everything else should be the same. Same filename, same paths, etc.
Now the problem is that I don't know how the debian/rules file should look like, to first build the myproject package in a folder and then build the myproject-debug with different CMake options (-DCMAKE_BUILD_TYPE=debug) in a different folder, so the filenames can and should stay the same.
There is this CMake tutorial in the Debian documentation, but this doesn't fit my requirements. Because in this tutorial everything will be built in only one folder, and in this one folder there are different files. Then different .install files will be used to copy the needed files to each package. But since I have the same binary filename for each package myproject and myproject-debug this tutorial does not really fit my needs.
I already have the following lines in my debian/rules file:
override_dh_auto_configure:
dh_auto_configure -- -DCMAKE_BUILD_TYPE=release
But how can I run two different builds with two different build types?
For example, something like this, to split it up:
override_dh_auto_configure_release:
dh_auto_configure -- -DCMAKE_BUILD_TYPE=release
override_dh_auto_configure_debug:
dh_auto_configure -- -DCMAKE_BUILD_TYPE=debug
And run both in different folders so I can add both folders to two different packages.
Or maybe there is even a better solution I cannot imagine yet?
I'm trying to compile custom ejabberd module which has multiple files in src/ and include/.
My module is conforming to https://github.com/processone/ejabberd-contrib guidelines and using ejabberdctl module_check produces "ok".
When I'm compiling and installing module with only one source, everything works as expected and ejabberdctl module_install installs module.
But when I have multiple source files, ejabberdctl module_install doesn't compile anything but file with exact name of my module.
I want to avoid manual compilation and rebar if possible. So my question is how to configure my module and where so that ejabberdctl module_install can build and install my module.
If that is not possible could you point me to some rebar script that can serve that purpose.
And finally can someone point me to some resource that explains why Erlang/OTP projects seem to have rigid file structure (I'm coming form c++ world and there IMO, tools and projects are way more forgiving in respect to file structure)
Details:
Erlang 18
ejabberd 15.11.120 (built from source)
if you install module by "ejabberdctl" you don't have to config in ejabberd.yml
other way is install module with source, you must copy file beam to lib/ejabberd... and config file config like README.TXT
I want to include the GNU Scientific Library (GSL) in my Cocoa app so that the user needs not installing it locally first. Like most GNU packages, it's the standard configure / make / make install routine. But this won't work:
./configure --prefix ~/libgsl
make
make install
Since the prefix is local to my computer. And neither is this:
./configure --prefix (path to build folder)/libgsl
make
make install
What I want is essentially the GSL being contained entirely in my application, and I can call its functions without the users downloading anything else.
I'm rather new to Xcode 4 and the build system for Clang/GCC, having coming over from .NET. Any help is much appreciated.
Assuming there is not a framework-style build of the library, the way this is typically done when bundling with 3rd party libraries is to build the package as normal, install it in /usr/local, and configure your project to include and link from there. Building is the easy part though.
The tricky part is bundling up the .app correctly. You need to add a custom build stage (after the others) which first copies all the dependent .dylib files into your app bundle's Frameworks folder (using the environment variables to help; see Xcode docs). Then you need to use install_name_tool to get the app binary to look in the framework dir (as the embedded soname still thinks it is in /usr/local). This part is very fiddly and not well documented.
I've just extracted this from a working project where I use GSL. Just add this as an extra build phase in your XCode project as a Custom Script:
# Framework folder for Example.app
FRAMEWORKS_DIR=${TARGET_BUILD_DIR}/Example.app/Contents/Frameworks
# Create path if it doesn't exsit
mkdir -p ${FRAMEWORKS_DIR}
# Find the original linked path for libgsl
GSLLIB=`otool -L ${TARGET_BUILD_DIR}/Example.app/Contents/MacOS/Example | grep libgsl | cut -d" " -f1`
GSLPATH=`dirname $GSLLIB`
# Copy the dylibs into your app
cp /usr/local/lib/lib{gsl,gslcblas}.0.dylib ${FRAMEWORKS_DIR}
# Update embedded paths
install_name_tool \
-change ${GSLPATH}/libgsl.0.dylib \
#executable_path/../Frameworks/libgsl.0.dylib \
${TARGET_BUILD_DIR}/Example.app/Contents/MacOS/Example
This should work with a simple substitution of your app name.
This is basically the same as what you need to do to build a standalone Qt app, so the docs here are very relevant:
Deploying Qt on the Mac
It is worth reading up on bundles, frameworks and packaging. For example:
Mac OS X Framework Reference
This post is also relevant:
How do I link libraries in Xcode 4?
Note that GSL is published under the GPL, so your app would need to be similarly published in order to respect the license. Shipping the source is necessary, but not sufficient for compliance.