Those were two key issues where BetterAWStats was not matching the capabilities of AWStats.
If you don't want to wait for the next version release, you can get the current version from CVS and enjoy the two new features.
Programmer's Note about XML Datafile support:
I first wrote a complete PHP XML-Parser which I then commented out. The reason for this is that AWStats has its data-map on the beginning, but in the XML version this points to the BEGIN_... text instead of the start of the xml data set. So it cannot be used to jump directly to a line in the file. Additionally, I could not find out how to make PHP read only a section of the file. So parsing the file with an XML parser took about 10 times longer than with the std method since it would always read the whole file. Further, XML has trouble parsing &-signs in URLs. On top of that, the SIDER section of the data is corrupt in AWStats since for some entries. the first number is not in its own <td>-segment but simply split with a space after the URL.
Example:
- Code: Select all
<tr><td>/forum/memberlist.php 2</td><td>7172</td><td>0</td><td>0</td></tr>
All this made it better to parse the data by simply removing the XML from the file instead of using a proper XML parser. I did not look at the PERL code of AWStats, but I strongly assume AWStats is not using a XML reader either.
So there is currently no advantage in switching to XML data files, it is even a bit slower than the normal datafile.
