• For
• Let
• Where
• Order
• Return

Using this basic model is a very easy to construct simple XQueries.

Other interesting XQuery tricks.

• You can get distinct values using fn:distinct-values
• You can create your own functions. (Use the local namespace for these functions. This namespace already exists in XQuery.)

Most of this day was hands on stuff that involved XQuery and eXist. I really MUST get eXist permenantly installed on a test server here. Although truthfully that isn’t likely to happen until sometime AFTER the madness of ALA. Honestly, I came away from the workshop thinking “Wow, I understand enough of this to actually DO something productive.” This is always a good feeling because I can go back to the resources I was looking at before and they make much more sense now. Also getting contact who I can ask questions about XML-related stuff is always a plus. I’m sure I’ll need them in the not so distance future as UH embarks on more digital library efforts and real Web Services.

• = – equal to (this can also mean node exists)
• != – not equal to (this can also mean node does not exist)
• < – less than
• > – greater than

contains() is also an interesting function to use to test if the text of a given node contains a string. The instructors didn’t talk about starts-with() which does something similar and I like to use as well.

We also learned about combining documents using the document() function. I’d never used this before. It was SO cool and helpful. What the document() function does is allow you to import content from another document into your stylesheet. So if you have a list of people in one xml files and bios in another xml file then you can use this function in your XSLT to get content from both xml files. To do this you get need to get the content from the other xml document. This is done using <xsl:variable name=”variable_name” select=”document(‘filename.xml’)/> This maps the xml document’s content to the variable_name. You can then access this content using the following XPath syntax $variable_name/xpath/to/node

A significant part of the afternoon was spent working on creating XSLTs from scratch. This was really good practice for me because the document we were asked to style was pretty complicated. There were lots of nested tags and tags with both nodes and text in them, which creates a little bit of a challenge.

The last part of the day was a discussion of how xml is being used in libraries today. Patrick Yott talked about METS how it can be used to model complex digital objects, and combine different types of metadata into a single record (descriptive metadata, technical metadata, and structural metadata [relationships between files]). METS can also be used to move files from one system to another.

Brown is using the following metadata standards for their digital objects.

• METS
• MODS
• MIX
• METSRights

Brown creates a lot of their MODS for their digital object from the EAD finding aids for the collections that the objects were digitized from. They also use an XLST to check to see if a created MODS record has all of the necessary (according to Brown) elements filled in.

Then Matthew Gibson talked about XSL-FO (Formating Objects) FO processors, Apache FOP. Up until this point I’ve been kind of terrified of trying to write an XSL-FO. However, Matthew did a good job a explaining the basics in a really short time (less than an hour). He also provided examples that we could download from the class site and play with later. XSL-FO seems a lot less intimidating now. I’m hoping to try out some of my newly acquired skills to do a transformation from our EAD to printed PDFs. This will really make our Special Collections folks happy.

The afternoon session focused on XSLTs. Most of this was review for me. However, there were a few little tricks that I learned. One of them involved how to get data into an HTML attribute. Let’s say you want to create an email link.

You can’t do this: <a href=”mailto:<xsl:value-of select=”email”/>”><xsl:value-of select=”email”/></a> – it isn’t valid. You can’t put angle brackets within angle brackets.

The way I was taught to solve this was quite cumbersome. So I’m not going to show it. Instead, what I learned in the class is that you can solve this problem by doing this:  <a href=”mailto:{email}”><xsl:value-of select=”email”/></a> . This is a simple slick way to get the job done. For more information about it check out section 5.6 of the XSLT 2.0 specification (it is section 7.6.2 in the XSLT 1.0 spec).

The refresher on predicates was also helpful to me because I often have a need for this type of node selection but it is something that my previous XML class didn’t focus enough on.

Today is more advanced XSLT in the morning and then XML in Libraries, eXist, and the beginnings of XQuery in the afternoon. I’ve super excited about the next two days because these topics are the ones I know the least about.

However, today’s discussion was just the opposite because most people are using XML Schema instead of DTDs today. I also got to play with <oXygen/> and see how it helps you to create schemas. The presenter didn’t talk too much about the GUI portion of <oXygen/> for schema creation. I sort of wish he did because I like the visual representation that the GUI offers. That isn’t to say I don’t think that people shouldn’t have to know how to right an XML Schema file from scratch. It is more that taking advantage of the GUI when possible is nice.

Tomorrow is more on schemas and then XSLT. I asked the course instructors if they could talk a tiny, tiny, bit about XML-FO because I have a real need for it. I have lovely EAD finding aids that I need to make print versions of and XML-FO is the way to go. So we will see. I also want to know more about XForms but will likely have to take a separate class for that.