Possible improvement: run code after building

#1

There are some tasks I like to do automatically after building, but currently this is only available on Windows, through FlashDevelop.

There’s no reason Lime can’t run a command of your choice after building, we just need to figure out how it should work.

My suggestion is to use Sys.command(). So if you typed <postBuild command="echo Hello world" />, then Lime would call Sys.command("echo", ["Hello", "world"]);.

For more examples, see the issue I opened:

Thoughts?

#2

What if I want to pass '‘Hello World’ as single argument instead of [“Hello”, “world”] ? And what about prebuild?

#3

Added <preBuild /> to the issue description.

As for single arguments, what should the syntax be?

<!-- escaped spaces -->
<postBuild command="echo Hello\ world" />

<!-- quotes -->
<postBuild command="echo 'Hello world'" />
<postBuild command='echo "Hello world"' />

<!-- special attributes -->
<postBuild command="echo" arg0="Hello world" arg1="and so on" />
#4

I like last variant:
<postBuild command="echo" arg0="Hello world" arg1="and so on" />

or maybe something like:

<postBuild cmd="echo">
    <arg>Hello world</arg>
    <arg>and so on</arg>
</postBuild>
1 Like
#5

I like options with nested elements. But for simple call Sys.comand("echo", ["Hello world"]) it could be just <postBuild cmd="echo" args="Hello world"/> , like <assets> which currently supports both nested elements and single line.

haxeflag, haxedef, but postBuild . May be postbuild ?

#6

I like it!

I think I prefer <postBuild cmd="echo Hello world" />, because that’s how you’d type it into the command line. (And I’d be happy to allow cmd as an alternative to command.)

Good point. I’ll update the issue.

#7

Bumping the thread because I finally got around to implementing it. How’s this look?