##// END OF EJS Templates
automation: schedule an EC2Launch run on next boot...
automation: schedule an EC2Launch run on next boot Without this, launching EC2 instances constructed from the AMI won't go through the normal EC2 instance launch machinery. This missing machinery does important things like set up network routes to use the instance metadata service and process any UserData. Since EC2Launch now runs on subsequent boots and UserData is processed, we needed to make setting of UserData conditional on bootstrapping mode. Differential Revision: https://phab.mercurial-scm.org/D7113
Gregory Szorc -
r43528:c09e8ac3 default
Show More
Name Size Modified Last Commit Author
/ contrib / packaging / inno
build.py Loading ...
mercurial.iss Loading ...
modpath.iss Loading ...
readme.rst Loading ...
requirements.txt Loading ...
requirements.txt.in Loading ...

Requirements

Building the Inno installer requires a Windows machine.

The following system dependencies must be installed:

Building

The build.py script automates the process of producing an Inno installer. It manages fetching and configuring the non-system dependencies (such as py2exe, gettext, and various Python packages).

The script requires an activated Visual C++ 2008 command prompt. A shortcut to such a prompt was installed with Microsoft Visual C++ Compiler for Python 2.7. From your Start Menu, look for Microsoft Visual C++ Compiler Package for Python 2.7 then launch either Visual C++ 2008 32-bit Command Prompt or Visual C++ 2008 64-bit Command Prompt.

From the prompt, change to the Mercurial source directory. e.g. cd c:\src\hg.

Next, invoke build.py to produce an Inno installer. You will need to supply the path to the Python interpreter to use.:

$ python3.exe contrib\packaging\inno\build.py \
    --python c:\python27\python.exe

Note

The script validates that the Visual C++ environment is active and that the architecture of the specified Python interpreter matches the Visual C++ environment and errors if not.

If everything runs as intended, dependencies will be fetched and configured into the build sub-directory, Mercurial will be built, and an installer placed in the dist sub-directory. The final line of output should print the name of the generated installer.

Additional options may be configured. Run build.py --help to see a list of program flags.

MinGW

It is theoretically possible to generate an installer that uses MinGW. This isn't well tested and build.py and may properly support it. See old versions of this file in version control for potentially useful hints as to how to achieve this.