ICP-2152: Java Technologies
XML
Mini-Project 2
Description
This mini-project is intended to cover the parsing, use, and creation of XML documents.
In this mini-project you will:
• create various applications to handle the BBC Weather RSS Feed;
• perform basic parsing and processing of XML data;
• extend your knowledge of Swing UIs;
• become familiar with and use Java’s API for XML Processing (JAX-P).
This Mini-Project is marked out of 100 points, and contributes 15% to your module mark.
Conventions
File New Project... are menu items to be selected, they may be right click menus or the top level
window menus.
Finish , Alt are either buttons or keys that need to be pushed.
App.java is the name of a file, or input that can be used verbatim.
L indicates a task that you will need to complete prior to moving to the next.
XML Processing
XML is short for eXtensible Mark-up Language, and Java has several families of APIs to handle this as
part of the base JDK. XML is a free-form, hierarchical data format - similar in looks to HTML. XML can
use DTDs (Document Type Definitions) and Schemas to dictate what tags and data is valid in various
parts of the document.
The most elementary operation with XML is to be able to read and understand the data contained
within the files. This is known as parsing the document. Java provides two main APIs to parse and XML
document. These are the Simple API for XML-Processing (SAX) or the Document Object Model API
(DOM). Both of these are found within the Java API for XML Processing (JAX-P) part of the JDK. DOM
parsers and their use are described in Big Java. For more information about SAX parsers see: http://
docs.oracle.com/javase/tutorial/jaxp/sax/parsing.html. To provide additional
help, there is an example SAX parser provided on Blackboard SAXParserExample which requires
both the Employee class and employees.xml.
For those of you know very little about XML you should begin by looking at Chapter 22 in Horstmann’s
Big Java (4th ed.). When reading this chapter make sure that you distinguish between XML and the
various Java APIs for processing XML documents. Also have a look at the Additional Notes in the
Appendix.
Finally, when creating your own XML documents use and IDE or text editor to provide maximum
support (e.g. syntax checking/highlighting, formatting, and code completion). Eclipse can be used to
help here, and will create XML documents in one of three ways:
Mini-Project 2 1
ICP-2152: Java Technologies
1. From a DTD;
2. From an XML Schema;
3. From a Well-Formed Template.
To access these functions, right click on the folder which should contain the new file and select New
Other... XML XML File . This will show the New XML File Wizard. To begin with, simply name your
file and choose Finish . This will open a new XML File in the XML Editor – this provides additional
support that the Java Source editor and plain file editor won’t.
The reason XML is used on the Internet is its’ universal and self- documenting nature. The description
of the data is contained in the hierarchy of elements in the document and (if present) in the schema
or type definition. This means that programs on different systems, written in different languages can
consume the same data and ‘understand’ it in the same way: whether the data is presented in the
same way and for the same purpose is neither here nor there.
Task 1 Consume the BBCWeather RSS Feed (15%)
RSS, short for Really Simple Syndication, is a type of XML data feed. It is popular with news, weather
and other current affairs sites, such as blogs, because it allows them to offer their content in an easy
to consume form. https://www.bbc.co.uk/weather/about/17543675 has details about
the RSS Feed service.
To begin, you will need to find the GeoID of a place to find the weather for. To do this, visit https:
//www.bbc.co.uk/weather and use the search bar at the top to select your desired location, for
example ‘Bangor, Gwynedd’. After clicking the link for your place, you will be taken to the observations
page. This page will have a URL similar to https://www.bbc.co.uk/weather/0/2656397.
The last set of numbers (2656397 in this case) is the GeoID. You will need to keep that safe for
later.
On the RSS Feeds page, the BBC give an example of the URL for the ‘Latest Observations’https://
weather-broker-cdn.api.bbci.co.uk/en/observation/rss/2643123. You would
replace the last seven digits (the GeoID) with the one for the location you are interested in. Visiting
the URL in Chrome will show the XML code natively, in Firefox the code will be shown or a download
popup depending on the version. Safari will ask you to open the News app.
We are interested in the code, which will look similar to:
1
2 3 xmlns:dc="http://purl.org/dc/elements/1.1/"
4 xmlns:georss="http://www.georss.org/georss" version="2.0">
5
6
7 https://www.bbc.co.uk/weather/2643123
8 Latest observations for Manchester from BBC Weather, including
weather, temperature and wind information
9 en
10 Copyright: (C) British Broadcasting Corporation, see http://www.bbc.
co.uk/terms/additional_rss.shtml for more details
11 Tue, 15 Jan 2019 08:00:00 GMT
12 2019-01-15T08:00:00Z
13 en
14 Copyright: (C) British Broadcasting Corporation, see http://www.bbc.
co.uk/terms/additional_rss.shtml for more details
15
16
17
Mini-Project 2 2
ICP-2152: Java Technologies
18 https://www.bbc.co.uk/weather/2643123
19 Temperature: 7C (44F), Wind Direction: South South Westerly, Wind
Speed: 4mph, Humidity: 81\%, Pressure: 1017mb, Rising, Visibility: --description>
20 Tue, 15 Jan 2019 08:00:00 GMT
21 https://www.bbc.co.uk/weather/2643123-2019-01-15
T08:00:00.000Z
22 2019-01-15T08:00:00Z
23 53.4809 -2.2374
24
25
26
Now that you have a URL and content, your task is to write a command- line client to retrieve the
document, parse the XML and display the current observation headline (the text highlighted in red).
The current headline is found in the
Some BBC style weather symbols shown in Figure 2.
Mini-Project 2 4
ICP-2152: Java Technologies
When there is no symbol that matches the description, you will need to make a decision on how best
to present the information to the user.
The images can be produced in a number of ways. For instance, you could build a local library of image
files by searching the Web. These images can then be used build instances of the class ImageIcon
which can then be used to construct a JLabel. Alternatively, you could use 2D graphics to produce
your own set of weather symbols. If you decide on the second option speak with the lecturer to
obtain further guidance.
L Once you have this client working, make sure to save the code unaltered so that it can be
assessed.
Task 4 Searching for Locations (20%)
The BBC uses the GeoNames ID system. In order to match names of specific locations to IDs
(e.g. Bangor, Gwynedd to 2656397), you will need to register with the GeoNames service at
http://www.geonames.org/login. Once your account details have been confirmed via
the e-mail, you will need to enable the ‘search’ web service at http://www.geonames.org/
manageaccount.
A description of the service can be found at http://www.geonames.org/export/
geonames-search.html.
One way to use the GeoNames search service is to supply a URL with search parameters to your web
browser. For example, the URL below will return a single search result, in English, for the ‘place’ or
text ‘london’ and user ‘eesa03’ (i.e. Steve Marriott):
http://api.geonames.org/search?q=london&maxRows=1&lang=en&username=
eesa03
This search returns;
1
2
3 5699
4
5 London
6 London
7 51.50853
8 -0.12574
9 2643743
10 GB
11 United Kingdom
12 P
13 PPLC
14
15
Using this web service amend your weather application so that instead of getting the user to supply a
lengthy URL he or she simply has to supply the name of the location, for which a weather forecast is
desired. Obviously, this means that you should alter the GUI to accommodate this modification.
Once the user has entered the location name, your program should pass it to the appropriate
GeoNames service. The service will then return an XML document, which can be parsed so that the
geonameId value can be extracted and used to build the required URL for the BB weather feed. For
example, the URL for London would become; https://weather-broker-cdn.api.bbci.
co.uk/en/observation/rss/2643743.
Mini-Project 2 5
ICP-2152: Java Technologies
You must remember to use your own username in the search URL supplied to GeoNames.
Additionally, the GeoNames service will return IDs for more places than the BBC can return weather
forecasts for, although forecasts are not limited to just the UK. When this happens the feed will return
an HTTP 404 – Not Found error. You will need to handle this event gracefully; an error message would
be appropriate.
Write a new version of your Weather Application to allow searches using the GeoNames API to find
the IDs before showing the weather for that location.
L Once you have this client working, make sure to save the code unaltered so that it can be
assessed.
Task 5 Additional Information from the Description (10%)
This exercise involves parsing and extracting the information provided in the description element
of the BBC Weather RSS feed. In particular you will need to recover the six pieces of information
highlighted in red below:
Temperature: 9C (48F), Wind Direction: South South
Westerly, Wind Speed: 19mph, Humidity: 87%, Pressure: 1018mb,
Rising, Visibility: Good
Having recovered this information you will need to find some suitable way of presenting it visually.
Marks will be awarded for use of advanced Swing components such as JTables, JTrees and JLists or
further Java2D drawing/image processing where appropriate. This aspect of the mini-project is free
form, however the assessor will ask questions regarding your design choices. This means that you
cannot simply decide to use the most complicated component possible to get good marks. The
choice must be appropriate for the information you are trying to display.
L Once you have this client working, make sure to save the code unaltered so that it can be
assessed.
Task 6 Serialising Search Data to XML (20%)
This final exercise requires you to serialise all of the search data, and retrieved GeoNames IDs back
into an XML file. The file needs to be written to as each search is completed so that the information
doesn’t need to be held in memory. You should use the Streaming API for XML (StAX) in Java to act as
both a reader (i.e. read from an XML document) and a writer (write to an XML document). You’ll find
Oracle’s tutorial for StAX at http://docs.oracle.com/javase/tutorial/jaxp/stax/
index.html.
The XML document you are going to create should conform to the basic structure shown below (the
elements emphasised in italics must be replaced with the actual values):
1
2
3 entered search term
4 true/false at GeoNames
5 filled in if found
6
7 ... repeated for each search completed ...
8
The element is repeated each time the user runs another search. You should ensure
that the file is updated after each write, so that your code can be tested completely. If the file you
Mini-Project 2 6
ICP-2152: Java Technologies
have chosen to use already exists, you can erase the contents of the file on each new execution of
your program. You will need to handle the fact that your program is running from a JAR and so will
not necessarily have a file system location to use. There is a suitable method in the java.io.File
class to assist in creating temporary files safely. You will also need to find a suitable way to inform
the user of the location of this file.
Submission
This must be submitted as a .zip file (no other archive formats will be accepted) of your entire
project folder. This will include your POM, source files, web files, as well as compiled and packaged
project.
This can be exported from Eclipse directly by right-clicking on the Project, selecting Export.Choose
the General > Archive File option, then select your project, ZIP as the archive type and enter a
name/location for the file and click Finish.
Useful Links
In addition to looking at Horstmman I strongly recommend the following two sites for more
information about XML. http://www.ibm.com/developerworks/xml/
http://www.w3schools.com/DTD/default.asp
Another important source of information is the W3 consortium site which maintains a section
devoted to XML. The home page is listed below:
http://www.w3.org/XML/
Perhaps the most useful area at W3 can be found at
http://www.w3.org/standards/xml/
This link will take you to an authoritative description of the XML and the various standards associated
with this markup language.
For an interesting article on XML design by Uche Ogbuji see
http://www.ibm.com/developerworks/xml/library/x-eleatt.html
For information about the various Java APIs associated with XML check out the following links:
http://docs.oracle.com/javase/tutorial/jaxp/index.htmlhttp://www.ibm.
com/developerworks/java/library/x-jaxp/
Appendix 1: XML Namespaces
Due to the fact that various XML authors may make use of the same tags to represent different
data (especially since quite often a word will have multiple possible meanings), naming collisions
can occur. In order to prevent such collisions, we typically declare our XML tags as belonging to a
namespace. For example, imagine the following XML document representing a university, which has
many courses and many students:
1
2
3
4 6
5 John Smith
Mini-Project 2 7
ICP-2152: Java Technologies
6 67
7
8
9 5
10 Computing Laboratory 2
11 30
12
13
Due to the fact that various XML authors may make use of the same tags to represent different
data (especially since quite often a word will have multiple possible meanings), naming collisions
can occur. In order to prevent such collisions, we typically declare our XML tags as belonging to a
namespace. For example, imagine the following XML document representing a university, which has
many courses and many students:
1
2
3
4 6
5 John Smith
6 67
7
8
9 5
10 Computing Laboratory 2
11 30
12
13
In the above XML document, there are two collisions - one between id tags and another between
name tags. In the first instance, our id and name tags are in reference to a student ID and a student
name whereas, in the second we are referring to a course ID and course name. This would likely cause
confusion when attempting to validate our XML documents against a DTD as it would appear (by
choosing the same name for these tags) that the id and name tags should store the same information
whereas in reality, this is not what we want. In order to distinguish between these, we can use
namespaces e.g.
1
2
3
4 6
5 John Smith
6 67
7
8
9 5
10 Computing Laboratory 2
11 30
12
13
In the above example, we have used two namespaces in order to ensure that our tags are now unique
in meaning – naming collisions no longer occur and we cannot confuse one tag for another. The
second line of the XML document, which would normally be used to declare the root element of an
XML document now also declares two namespaces – student and course. We assign each namespace
a URL purely in order to ensure that these namespaces are unique. The theory is that since we are
the only one to which a particular URL belongs, by using our URL as part of our namespace, this
should ensure that our namespace is truly unique – using the Bangor University is probably not the
best example since in theory every student / staff member of the university might make use of this
Mini-Project 2 8
ICP-2152: Java Technologies
domain in order to define a namespace so it would probably be better to make use of a more personal
domain name should you happen to own one.
Whilst URLs are very commonly used in practice in order to make namespaces unique, we are in
fact allowed to use any URI i.e. we are not limited to URLs. Notice that http://www.bangor.
ac.uk/xmlns-student and http://www.bangor.ac.uk/xmlns-course aren’t actually
URLs which exist, the URL is included as part of the namespace purely for the purposes of making
it unique, for the same reason that Java packages are also typically named after (reversed) domains
e.g. uk.ac.bangor. Having declared our namespaces, we simply prefix each tag with the namespace
to which it should belong.
Mini-Project 2 9