Bits and bites from Pål Brattberg

Found this gem of knowledge for creating a comma separated list with XSLT without having to resort to the monstrous hacks I’ve used before.

My XML file is looking like this:

<message id="Login" initiator="client" system="lobby">
    <description>This messages is sent from the client when the client wants to login.</description>
    <request>
        <argument name="loginname" type="&string;" example="myUserName" maxsize="40"/>
        <argument name="password" type="&string;" example="myP4Ssw0rd!" maxsize="40">
            <description>Yes, this should be encrypted. Actual algorithm not yet
            decided.</description>
        </argument>
        <argument name="macaddress" type="&string;" example="09:00:07:A9:B2:EB">
            <description>We may wish to include even more information about the client PC at login,
                but for now see the CheckVersion message.</description>
        </argument>
    </request>
    <response> &frgUserId; <argument name="loginserverid" type="&integer;" example="12">
            <description>This is the id for the loginserver the client has been assigned
            to.</description>
        </argument>
        <argument name="token" type="&string;" example="33e24aeb44d25327" maxsize="2048"
            required="false">
            <description>This is a string for the unique token that identifies this particular
                client's login.</description>
        </argument>
        <argument name="errornumber" type="&integer;" example="0">
            <description>This should be >0 in case of an error. These error codes are
                described in another document.</description>
        </argument>
    </response>
</message>

Now, I want to generate source code from this using XSLT. In one instance I generate method signatures that need all request arguments as parameters for the method. I then have a snippet of XSL generate the code like this (simplified):

<xsl:for-each select="//message">
    <![CDATA[public void on]]><xsl:value-of select="@id"/>(<xsl:if test="request != ''">
        <xsl:for-each select="request/*">
            <xsl:value-of select="@type"/>
            <xsl:value-of select="concat(' ', @name)"/>
            <xsl:if test="position() != last()">, </xsl:if>
        </xsl:for-each>
    </xsl:if>);</xsl:for-each>

The beauty is of course in </xsl:if>);</xsl:for-each> which simply tests if this position is the same as the last position. Simple and elegant!

Now, if you are using XSLT 2.0 you could go one step further and skip the entire inner xsl:for-each and do it in one line. This solution is not quite as simple in this particular example considering that I need to concatenate several attributes for each item in the list, but it’s still doable.

<xsl:for-each select="//message">
    <![CDATA[public void on]]><xsl:value-of select="@id"/>(<xsl:if test="request != ''">
        <xsl:value-of select="for $s in (request/*) return (concat($s/attribute::type, ' ', $s/attribute::name))" separator=", "/>
    </xsl:if>);</xsl:for-each>

Lets say you wanted something simple, such as a list of all the message id’s defined in the XML file, this would be done like this in 1.0:

<xsl:for-each select="//message">
    <xsl:value-of select="@id"/>
    <xsl:if test="position() != last()">, </xsl:if>
</xsl:for-each>

And the 2.0 equivalent is:

<xsl:value-of select="//message/@id" separator=", "/>

In this example the 2.0 solution is much cleaner in my mind. But choose the one you like the most!

Got the following juice link from my colleague Johan Ekblad:

A Quick Primer On Unicode and Software Internationalization Under Linux and UNIX

This page provides a quick summary of information with links to other URLs regarding using Unicode for multilingual internationalization projects on Linux and other UNIX-based operating systems. If you would like to be able to use more than one language on your Linux/UNIX computer but haven’t completely figured out how to do that yet, then you should read this page.

Go ahead and read it, has some good tips on what to do if you get in trouble with foreign characters, i18n, l10n or some other buzzword.

In my current project, I do most of the database modelling as well as implementation of the database layer and since the project includes user registration of potential users from around the world we need some tables for keeping track of available countries.

While searching the web for previous work (don’t reinvent the wheel my friend) I came upon a nice resource by Bill Rhodes at http://27.org/isocountrylist/. This is a great resource for codes and names for both countries and states, complete with a perl-script to keep it synchronized with the ISO 3166 standard. Sweet!

That was a big help of course, now all I needed was a list of currency codes and names. I couldn’t locate a nicely formatted text file of ISO 4217, but found this web based list of currencies that I could build upon.

I only need two codes for now, but if I would’ve found the corresponding text file at ISO.org I would’ve/could’ve adapted Bill’s perl to also fix the currency table.

My solution was dead simple:
DROP TABLE IF EXISTS `currency`;
CREATE TABLE `currency` (
iso3 CHAR(3) NOT NULL PRIMARY KEY,
printable_name VARCHAR(80) NOT NULL,
numcode SMALLINT,
country_iso3 CHAR(3)
);
INSERT INTO currency VALUES ('USD','US Dollar','840', 'US');
INSERT INTO currency VALUES ('SEK','Swedish Krona','752','SE');


If I get around to do the script, I will send it to Bill so he may keep it there as well.