View source for Internationalization of Haskell programs using gettext

Most common in the GNU world approach to internationalization (i18n) of software is to use [http://www.gnu.org/software/gettext/ GNU gettext] utilities. In this tutorial we will create simple "Hello world" program, with multilingual support.

==Prepare program to internationalization==

Consider we want to make the following program multilingual (file '''Main.hs'''):

<haskell>
module Main where
    
import IO 

main = do
  putStrLn "Please enter your name:"
  name <- getLine
  putStrLn $ "Hello, " ++ name ++ ", how are you?"
</haskell>

First of all, wrap all strings you want to translate in function <hask>__</hask>:

<haskell>
module Main where
    
import IO 

__ = id

main = do
  putStrLn (__ "Please enter your name:")
  name <- getLine
  putStrLn $ (__ "Hello, ") ++ name ++ (__ ", how are you?")
</haskell>


We will return to the definition of <hask>__</hask> a bit later, for now leave this function as a no-op (<hask>id</hask>).

==Translate==

The next step is to generate a POT file (a template which contain all strings needing translation). For Python, C, C++ and Scheme languages there is the xgettext utility, but it doesn't support Haskell. On [[Hackage]] you can download [http://hackage.haskell.org/cgi-bin/hackage-scripts/package/hgettext hgettext] library and utilities, which process haskell source files in the same way as xgettext does for C/C++ files:

<tt> cabal install --global hgettext </tt>

Now run this from the directory where your project is:

<tt>hgettext -k __ -o messages.pot Main.hs</tt>

This gathers all strings marked by the function <tt>__</tt> from the file <tt>Main.hs</tt> and writes everything to <tt>messages.pot</tt>.

Now look at the resulting pot file:

<tt>

<pre>
# Translation file
 
msgid ""
msgstr ""
 
"Project-Id-Version: PACKAGE VERSION\n"
"Report-Msgid-Bugs-To: \n"
"POT-Creation-Date: 2009-01-13 06:05-0800\n"
"PO-Revision-Date: YEAR-MO-DA HO:MI+ZONE\n"
"Last-Translator: FULL NAME \n"
"Language-Team: LANGUAGE \n"
"MIME-Version: 1.0\n"
"Content-Type: text/plain; charset=UTF-8\n"
"Content-Transfer-Encoding: 8bit\n"

#: Main.hs:0
msgid "Please enter your name:"
msgstr ""

#: Main.hs:0
msgid "Hello, "
msgstr ""

#: Main.hs:0
msgid ", how are you?"
msgstr ""
</pre>
</tt>


We are interested in the bottom part of this file (starting from <tt>'#: Main.hs:...'</tt>). Here we can see pairs of lines: <tt>msgid</tt> and <tt>msgstr</tt>: <tt>msgid</tt> is the original text from the code, and <tt>msgstr</tt> is the translated string. Each language should have its own translation file. I will create two translations, German and English.

To create a PO file for specific locale we should use the <tt>msginit</tt> utility:

To generate the German translation file template run:

<tt>msginit --input=messages.pot --locale=de.UTF-8</tt>

And for the English translation run:

<tt>msginit --input=messages.pot --locale=en.UTF-8</tt>

If we look at the generated files (<tt>en.po</tt> and <tt>de.po</tt>), we will see that the English translation is completely filled, we have only to edit the German PO file. So fill it with the following strings:

<tt>
<pre>
#: Main.hs:0
msgid "Please enter your name:"
msgstr "Wie heißen Sie?"

#: Main.hs:0
msgid "Hello, "
msgstr "Hallo, "

#: Main.hs:0
msgid ", how are you?"
msgstr ", wie geht es Ihnen?"
</pre>
</tt>

==Install translation files==

Now we have to create the directories where these translations should be placed.  By default all translation files are placed on <tt>/usr/share/locale/</tt> folder, but we are free to select different places. Run:

<tt>mkdir -p {de,en}/LC_MESSAGES</tt>

It will create two directories in the current directory, <tt>de</tt> and <tt>en</tt>, that contain <tt>LC_MESSAGES</tt>. Now use <tt>msgfmt</tt> tool to encode our <tt>po</tt> files to <tt>mo</tt> files (binary translation files):

<tt>
<pre>
msgfmt --output-file=en/LC_MESSAGES/hello.mo en.po
msgfmt --output-file=de/LC_MESSAGES/hello.mo de.po
</pre>
</tt>

==Enable internationalization in the code==

As the final step we have to modify the code to support the internationalization:

<haskell>
module Main where
    
import IO 
import Text.I18N.GetText
import System.Locale.SetLocale
import System.IO.Unsafe

__ :: String -> String
__ = unsafePerformIO . getText

main = do
  setLocale LC_ALL (Just "") 
  bindTextDomain "hello" "." 
  textDomain "hello" 

  putStrLn (__ "Please enter your name:")
  name <- getLine
  putStrLn $ (__ "Hello, ") ++ name ++ (__ ", how are you?")
</haskell>


Here we added three initialization commands:

<haskell>
setLocale LC_ALL (Just "") 
bindTextDomain "hello" "." 
textDomain "hello" 
</haskell>


The first one (you'll have to download [http://hackage.haskell.org/cgi-bin/hackage-scripts/package/setlocale setlocale] package to enable this function) sets the current locale to a default value. The next two functions tell <tt>gettext</tt> to take the '''"hello.mo"''' message file from the locale directory (I set it to ".", but, in the general case, this directory should be passed from the package configuration).

The final step — define function <hask>__</hask>. It simply calls <hask>getText</hask> from the module <hask>Text.I18N.GetText</hask>, but its type is <hask>String -> IO String</hask> so here we used <hask>unsafePerformIO</hask> to make it simpler to call.

==Run the program==

Now you can build and try this program in different locales:

<tt>
<pre>
user> ghc --make Main.hs
[1 of 1] Compiling Main             ( Main.hs, Main.o )
Linking Main ...

user> LOCALE=en_US.UTF-8 ./Main
Please enter your name:
Bond
Hello, Bond, how are you?

user> LOCALE=de_DE.UTF-8 ./Main
Wie heißen Sie?
Bond
Hallo, Bond, wie geht es Ihnen?

user>
</pre>
</tt>

==Distribute internationalized cabal package==

TBD