Re^5: use strict and TokeParser

No problem. If you happen to recall the bug and can tell me the steps to reproduce it, I'll be happy to fix it and add a test.

Cheers,
Ovid

New address of my CGI Course.

Comment on Re^5: use strict and TokeParser

Replies are listed 'Best First'.
Re^6: use strict and TokeParser by runrig (Abbot) on Jun 03, 2005 at 18:32 UTC
Using HTML::LinkExtractor v.11 from backpan, and the bug previously mentioned, I reproduced the bug, and could not find anything wrong with HTML::TokeParser::Simple(v3.13). I did however find this in HTML::LinkExtractor: `## In case we got nested tags if(@TEXT) { $TEXT[-1]->{_TEXT} .= $T->as_is; }` [download] I have some idea of what the purpose of the code above was, but it had the effect of setting the current "text" of the link (even the top level non-nested link tag since the @TEXT array was non-empty and the _TEXT contained an empty string) to the entire <a href="..."> tag (and then later the actual text is appended).	[reply] [d/l]

Replies are listed 'Best First'.

Re^6: use strict and TokeParser
by runrig (Abbot) on Jun 03, 2005 at 18:32 UTC

## In case we got nested tags
if(@TEXT) {
    $TEXT[-1]->{_TEXT} .= $T->as_is;
}
[download]