Reading Specific Parts of an XML Document

[Jordan]

I'm using an xml document as the source for my site's navigation. I have it
working but am curious if I've chosen the best method (performance wise) of
reading this xml document. Becuase this xml file has my site's complete
navigation, I need to "jump around" in it to get to the data I need based on
the page. I'm using the Skip() method of the XmlReader class to do this.

Is this the best way to move forward in an xml document?

I also have another question too now that I'm thinking about it - what would
be the best way to cache my navigation as it won't change often at all?

tia, j

 

[Cowboy \(Gregory A. Beamer\)]

Consider adding the metadata for navigation, if not there already, as nodes.
Then, you can use XPath to prune out the nodes you want. For example:

<SideMenu>
<LinkName>Policies</LinkName>
<LinkURL>policies.aspx</LinkURL>
<Section>Home</Section>
</SideMenu>

NOTE: The XML here is DataSet style, but the principle is still the same.

With XPATH, you can set up a query that only finds those side menu items
that are in section Home. The reason I use DataSet XML for menus is we are
aiming for eventually moving the menus to a database, with an admin utility
for all of our applications. By starting with XML that is DataSet oriented,
I can easily change to the database, when finished, without rewriting the
bind.

--
Gregory A. Beamer
MVP; MCP: +I, SE, SD, DBA

**********************************************************************
Think Outside the Box!
**********************************************************************

 

[Jose Marcenaro]

Answering both question in one:

If the navigation file is global to all users, store it as an Application[]
variable or -to avoid serialization - as a static/shared property of a
class.
In any case, if the navigation xml is cached you may keep it as an
XmlDocument object, and use the XPath query syntax for a fast search.

Going a little further, you may encapsulate the navigation logic as
static/shared members of a class (which also holds the xml as a static
private property). That way the navigation file access would be contained
within that class and would be easily changed with no impact on the
application

Regards
Jose.

 

[Jordan]

Ok - I haven't found in asp.net that you can use XPath syntax when using
something like
new xmlReader(MyXmlFile). I know there's an XQuery component that I can
download, but I was hoping to stay away from another component at this time.

right now, my xml file is not DataSet style. Here's a snippet
<navigation>

<node area="Courses" root="/courses">

<page url="/courses" headeronly="false" display="expand">Browse Courses

<subpage url="/courses/browsebysubject.asp">by Subject</subpage>

<subpage url="/courses/browsebynumber.asp">by Course Number</subpage>

</page>

<page url="/courses/index.asp">Course Index</page>

<page url="/courses/basics.asp">Course Basics</page>

</node>

<node> .... </node>

What I'm doing is just using the While (xmlReader.Read()) and going through
it on my own, not using any specific DataBind to a control, but I am using:

MyDiv.InnerHtml = MyCustomStringBuilderFromMyWhileLoop.ToString();

Is your method better (have an article maybe)??

 

[Jordan]

Jose,

Do you mean I can do something like this?

Application["MyXmlFile"] = new xmlReader(MyXmlFile);

My reply to "Cowboy" earlier said that I didn't know you could use XPath in
..NET exactly, but use XQuery for example. Seems like I'm wrong about this.

Going a little further, you may encapsulate the navigation logic as
static/shared members of a class (which also holds the xml as a static
private property). That way the navigation file access would be contained
within that class and would be easily changed with no impact on the
application

This is kinda what I was trying to accomplish. I was going to compile my
nav.dll and in my pages, just have my <div id="myNav"> and use
MyVav.InnerHtml = MyDLLsCustomHTMLString. That way, I can change the XmlDoc
and have my navigation change w/o impacting the application too.

Answering both question in one:

If the navigation file is global to all users, store it as an Application[]
variable or -to avoid serialization - as a static/shared property of a
class.
In any case, if the navigation xml is cached you may keep it as an
XmlDocument object, and use the XPath query syntax for a fast search.

Going a little further, you may encapsulate the navigation logic as
static/shared members of a class (which also holds the xml as a static
private property). That way the navigation file access would be contained
within that class and would be easily changed with no impact on the
application

Regards
Jose.

I'm using an xml document as the source for my site's navigation. I have it
working but am curious if I've chosen the best method (performance wise) of
reading this xml document. Becuase this xml file has my site's complete
navigation, I need to "jump around" in it to get to the data I need

based

on
the page. I'm using the Skip() method of the XmlReader class to do this.

Is this the best way to move forward in an xml document?

I also have another question too now that I'm thinking about it - what would
be the best way to cache my navigation as it won't change often at all?

tia, j

 

[Jose Marcenaro]

Jordan,

Do you mean I can do something like this?
Application["MyXmlFile"] = new xmlReader(MyXmlFile);

I didn't think of keeping an open reader; I actually meant something like
xmlDoc = new XmlDocument();
xmlDoc.load( MyXmlFile);
Application["MyXmlDoc"] = xmlDoc;
or rather
MyNavigationClass.XmlDoc = xmlDoc; // static member

You could also use the Cache object, which also allows for refreshing the
data when needed

Hope this helps
Jose.

Jose,

Do you mean I can do something like this?

Application["MyXmlFile"] = new xmlReader(MyXmlFile);

My reply to "Cowboy" earlier said that I didn't know you could use XPath in
.NET exactly, but use XQuery for example. Seems like I'm wrong about this.

Going a little further, you may encapsulate the navigation logic as
static/shared members of a class (which also holds the xml as a static
private property). That way the navigation file access would be contained
within that class and would be easily changed with no impact on the
application

This is kinda what I was trying to accomplish. I was going to compile my
nav.dll and in my pages, just have my <div id="myNav"> and use
MyVav.InnerHtml = MyDLLsCustomHTMLString. That way, I can change the XmlDoc
and have my navigation change w/o impacting the application too.

Answering both question in one:

If the navigation file is global to all users, store it as an Application[]
variable or -to avoid serialization - as a static/shared property of a
class.
In any case, if the navigation xml is cached you may keep it as an
XmlDocument object, and use the XPath query syntax for a fast search.

Going a little further, you may encapsulate the navigation logic as
static/shared members of a class (which also holds the xml as a static
private property). That way the navigation file access would be contained
within that class and would be easily changed with no impact on the
application

Regards
Jose.

I'm using an xml document as the source for my site's navigation. I

have
it

working but am curious if I've chosen the best method (performance

wise)
of

reading this xml document. Becuase this xml file has my site's complete
navigation, I need to "jump around" in it to get to the data I need

based

on
the page. I'm using the Skip() method of the XmlReader class to do this.

Is this the best way to move forward in an xml document?

I also have another question too now that I'm thinking about it - what would
be the best way to cache my navigation as it won't change often at all?

tia, j