VirtualRunEnv

Warning

These tools are still experimental (so subject to breaking changes) but with very stable syntax. We encourage the usage of it to be prepared for Conan 2.0.

The VirtualRunEnv generator can be used by name in conanfiles:

conanfile.py
class Pkg(ConanFile):
    generators = "VirtualRunEnv"
conanfile.txt
[generators]
VirtualRunEnv

And it can also be fully instantiated in the conanfile generate() method:

conanfile.py
from conans import ConanFile
from conan.tools.env import VirtualRunEnv

class Pkg(ConanFile):
    settings = "os", "compiler", "arch", "build_type"
    requires = "zlib/1.2.11", "bzip2/1.0.8"

    def generate(self):
        ms = VirtualRunEnv(self)
        ms.generate()

When the VirtualRunEnv generator is used, calling conan install will generate a conanrunenv .bat or .sh script containing environment variables of the run time environment.

The launcher contains the runtime environment information, anything that is necessary in the environment to actually run the compiled executables and applications. The information is obtained from the self.runenv_info and also automatically deduced from the self.cpp_info definition of the package, to define PATH, LD_LIBRARY_PATH, DYLD_LIBRARY_PATH and DYLD_FRAMEWORK_PATH environment variables.

This generator will create the following files:

  • conanrunenv-release-x86_64.(bat|sh): This file contains the actual definition of environment variables like PATH, LD_LIBRARY_PATH, etc, and runenv_info of dependencies corresponding to the host context, and to the current installed configuration. If a repeated call is done with other settings, a different file will be created.

  • conanrun.(bat|sh): Accumulates the calls to one or more other scripts to give one single convenient file for all. This only calls the latest specific configuration one, that is, if conan install is called first for Release build type, and then for Debug, conanrun.(bat|sh) script will call the Debug one.

After the execution of one of those files, a new deactivation script will be generated, capturing the current environment, so the environment can be restored when desired. The file will be named also following the current active configuration, like deactivate_conanrunenv-release-x86_64.bat.

Let’s see an example on how to add an environment variable to the runenv_info and get its value later in the consumer side using the conanrun launcher:

conanfile.py
from conan import ConanFile

class HelloConan(ConanFile):

    def package_info(self):
        self.runenv_info.define("MYVAR", "My value!")
test_package/conanfile.py
from conan import ConanFile

class HelloTestConan(ConanFile):
    # VirtualBuildEnv and VirtualRunEnv can be avoided if "tools.env.virtualenv:auto_use" is defined
    # (it will be defined in Conan 2.0)
    generators = "VirtualRunEnv"

    def test(self):
        self.run("echo $MYVAR", env="conanrun")  # Unix-style

As we already said above, the conanrun launcher contains the runtime environment information, so let’s run a conan create . hello/1.0@ and check the console output that should show something like this:

....
Configuring environment variables
My value!

Constructor

def __init__(self, conanfile):
  • conanfile: the current recipe object. Always use self.

generate()

def generate(self, scope="run"):

Parameters:

  • scope (Defaulted to run): Add the launcher automatically to the conanrun launcher. Read more in the Environment documentation.