Content Rotator

The Content Rotator is a simple but useful ASP object built with Microsoft's ActiveX Template Library 1.1 (ATL).

The object was created with the the ATL COM AppWizard. It contains four methods that were added by hand to the IDL file (ContRot.idl), the object's header file (RotObj.h), and the object's C++ implementation file (RotObj.cpp):

HRESULT OnStartPage(IUnknown* pUnk)
HRESULT OnEndPage()
HRESULT ChooseContent(BSTR bstrDataFile, BSTR* pbstrRetVal)
HRESULT GetAllContent(BSTR bstrDataFile)

OnStartPage and OnEndPage

OnStartPage and OnEndPage are two optional methods that ASP calls on an object whenever a page is opened or closed by the user's Web browser. The OnStartPage method can use the IScriptingContext interface to retrieve a pointer to an interface pointer of a built-in object that enables a component to access that object's collections, methods, and properties.

ChooseContent

ChooseContent reads a collection of weighted entries from bstrDataFile and returns one entry in pbstrRetVal. Each entry is a piece of HTML text. It can be used as:

a tip-of-the-day generator
a random advertisement billboard
a poor man's load-balancing scheme for selecting email names of Tech Support people.

A data file for the Content Rotator looks like this:

%% // Tech Support people
<a href="mailto:jsmith">John Smith</a>, Tech Lead
%%
<a href="mailto:bkidd">Billy (the) Kidd</a>, Trouble Shooter
%% #3	// Jane is full-time, so make her the likeliest to be chosen
<a href="mailto:jdoe">Jane Doe</a>, Product Support
%% // Scare 'em away.  Bwah ha ha!
Sorry.  Tech Support<br>
is<br>
on<br>
<b>Strike!</b>

It contains a number of weighted entries. The entries are HTML text. Each entry starts with one or more lines beginning with %%. If the %% line doesn't have anything on it (as in the one for Billy Kidd), then the entry's weight is one. If there's a number preceded by a # (as in the Jane Doe example), then that's the weight. Comments on the %% line start with //.

The probability of an entry's being chosen by the Content Rotator is its weight divided by the sum of the weights of all the entries (usually 1/n).

GetAllContent

GetAllContent reads a collection of weighted entries from bstrDataFile and writes all the entries back to the user's browser, each one separated by a horizontal rule (an <HR> tag), so that it can be proofread.

Demos

Click here to run a demonstration of the ChooseContent method.

Click here to run a demonstration of the GetAllContent method.