[Htmlparser-user] Node object and line number

[Marc N. <ma...@ke...>]

Hello,

I just thought I'd start out by thanking everyone who has worked on the =
htmlparser project.  I'm been using it for only a few days now but the =
functionality it provides has saved me amazing amounts of work.  So far =
I have found it very easy to integrate into my project.

I am using the htmlparser library for what I'm guessing is a less =
traditional application.  I've integrated it into a custom servlet =
filter which takes a processed JSP page and parses it for "custom" tags =
which I've defined.  Using custom scanner and tag, I'm able to replace =
my "custom tags" with appropriate HTML/Javascript in the toHtml() method =
for each tag.  However, I'd like to add some validation to my code to =
ensure certain constraints are observed, such as certain tags which =
REQUIRE a "name" attribute to be defined.  I've done this easily enough =
by adding a "verify()" method to my custom tags and throwing a =
ParserException if a constraint is violated.

However, just throwing an exception does not help the webpage developer =
determine where the problem is in the HTML.  What would REALLY help me =
is if the Node object had a method on it called something like =
getLineNumber() which returned the line number at which that node was =
parsed.

I've looked at the source code and this seems feasible.  The NodeReader =
class keeps track of the current line number as it finds nodes in the =
HTML.  Maybe the constructor for a Node() object could take in one more =
argument, the lineNumber, so that it could expose that lineNumber in a =
public method.

Does this sound like a hairbrained idea?  Has this ever come up before?

Thanks again,
Marc Novakowski