In many cases we need to customize the features of the software, then we need to use the source code installation. So, there are many tools to compile the source code, then, how do we know what other people are using the tool it. In fact, I do not know (*^▽^*).
Then this blog is mainly to write C code source installation.
Don't say anything, just go ahead and summarize the four steps of the source installation:
The first step:./configure: "Of course, we must also unzip the source package before this step" (1) Pass the parameter through the option, specify enable feature, install path, etc., refer to User's designation and makefile.in file generation makefile when executing (2) Check the dependent external environment Step Two: Make: Build the application according to the makefile file; Step Three: Make install Fourth step: Post-installation configuration options "In fact, the next two steps are simple, the focus is the first step and the fourth step" |
The above is just a small sum up, then the concrete how to achieve, GG
1. How do we get the source package
Official self-built site: Apache.org (ASF) mariadb.org ... Code hosting: SourceForge Github.com Code.google.com |
2, is a big problem, compile C source code
Premise: Provide development tools and development environment
Development tools: Make, GCC, etc.
Development environment: Development Library, header file
GLIBC: Standard Library
To provide development components through "package groups"
CentOS 6 primarily installs these two development package components: "Development Tools", "Server Platform development"
First Step: Configure script
We have to provide some parameters to the script: such as specifying the installation location, specifying the Enable feature, and so on, we can use the command:./configure--help to get the options it supports.
Option Categories: Configuration:
Installation Directories: Installation path setting --prefix=/path/to/somewhere: Specifies the default installation location; default is/usr/local/ --sysconfdir=/path/to/somewhere: Configuration file installation location; System types: Target platform Architecture "NOTE: This option is only used when cross-compiling" Optional Features: Optional --disable-feature --enable-feature[=arg] Optional Packages: A dependent package that defines a dependency --with-package[=arg] --without-package Some Influential environment variables: things that may affect environment variables |
Step two: Execute the "make" command
Step three: Execute "make install"
Here we can enable, now to our installation path down to boot (because we have not configured the program) Here's a quick start. To our installation location, execute the start command: e.g../apachectl Start |
Fourth step: Configuration options after successful installation:
(1) Export the binary program directory to the PATH environment variable, "If you do not export, we can only use absolute path to access him." Edit configuration file:/etc/profile.d/name.sh Create a script file Add an environment variable inside: Export path=/path/to/bin: $PATH
(2) Export Library file path Edit/etc/ld.so.conf.d/name.conf Create a profile to add a new library file in the same directory as this file To have the system regenerate the cache: command: ldconfig [-v]
(3) Export header file Method One: Go to the installation directory and copy the include header file to the/usr/include directory Method Two: Implement based on the link way LN-SV source file Soft Connect file (4) Export Help manual Edit/etc/man. Config file, add a Manpath parameter inside and indicate the path. |
Linux Software Management src source installation compile