So finally I decided to run a blog. And yes, you're reading my first blog post ever!
This afternoon I decided to give a whirl to Go Programming Language. I don't know yet so much about it but I'm intrigued by some concepts like goroutines and efficient GC. Moreover, I was amazed by the very first impact with the language: the building process is foolproof (at least on my Ubuntu box) so I decided to document the process as a tribute to this neatness. It's an example of how things should be done in computer programming. Moreover, it's a good starting point for my first blog post ever, isn't it?
Assuming you're on a debian based system you need - first of all - to install the toolchain to building the language:
sudo apt-get install bison gcc libc6-dev ed gawk make
And, if you decide to clone the mercurial repository, you need mercurial of course!
sudo apt-get install mercurial
Now that the prerequisites are satisfied let's configure our build. Here's the neatness of the whole process. You setup a bunch of environment variables and you're done! So for example, assuming that we wish to:
all we need is to export the following variables:
export GOROOT=$HOME/go export GOARCH=386 export GOOS=linux
And we can save them in .bashrc if we want a persistent configuration.
The installation process assumes that $HOME/bin folder exists and it places binary files in it. Let's meet this last prerequisite:
mkdir $HOME/bin
You can change binaries' folder setting the $GOBIN variable. In order to run the executables, don't forget to update your $PATH variable appending $HOME/bin.
export PATH=$PATH:$HOME/bin
Well, now let's go cloning the repo and building the whole stuff:
hg clone -r release https://go.googlecode.com/hg/ $GOROOT cd go/src ./all.bash
At this point the installation process assumes that $HOME/bin folder exists and it places binary files in it (you can change folder name setting the $GOBIN variable). And finally let's test our build starting the compiler:
$ 8g flags: -I DIR search for packages in DIR -d print declarations -e no limit on number of errors printed -f print stack frame structure -h panic on an error -o file specify output file -S print the assembly language -w print the parse tree after typing -x print lex tokens
And that's all, no autogen.sh, no configure scripts to run! And is even more beautiful because we are actually cross-compiling! That is, we could produce builds for any other supported OS/ARCHs by simply changing the values of the environment variables above! Can't wait to test on my shining new beagleboard when it will arrive! :D
For further readings see: