Bean properties - case convention anomaly

[chris leonardi]

Hello -
I'm running into a really odd problem on Tomcat 3.2.1. Here goes: I've got a bean. The bean has a bunch of properties. Some of the fields look like this:
...
protected String aFieldName;
protected String bFieldName;
protected String cFieldName;
...
The corresponding accessor methods look like this:
...
public void setAFieldName(String v)
{
this.aFieldName = v;
}
public String getAFieldName()
{
return this.aFieldName;
}
...
Pretty straightforward stuff, it would seem. However, when using the standard jsp:setProperty and jsp:getProperty tags for the bean's properties, it appears the the JSP engine's introspection "breaks down" - it can't seem to locate the appropriate accessor method (throws "no get method for property 'aFieldName' found in bean ....". I get essentially the same result using Jakarta Struts tag extensions to manipulate the bean. Interestingly enough, I can access the method's directly using scriptlet code ("<% String foo = myBean.getAFieldName(); %>" ).
Now here's where it gets really bizare. Change the field names: instead of having a lower-case first position character, followed by an upper-case second position character, add in another lower-case character. So:
...
protected String aaFieldName;
...
And modify the accessor methods accordingly:
public void setAaFieldName(String v)
{
this.aaFieldName = v;
}
public String getAaFieldName()
{
return this.aaFieldName;
}
...
Now it works. Can anyone else reproduce this phenomenon - either on Tomcat or another JSP engine? Am I missing something? Any ideas would be much appreciated - while I'd like to simply rename the fields and put the whole issue to rest, the bean properties are being defined by a client's design documents. Thanks,
Chris
*** update ***
for anyone who's interested, it looks like this is a documented "bug" in tomcat 3.2.1: http://nagoya.apache.org/bugzilla/show_bug.cgi?id=322
[This message has been edited by chris leonardi (edited March 16, 2001).]