Hi Folks,
Its time again, another integration release is out. Check =
http://htmlparser.sourceforge.net.
So whats new? Major API change - the parser now has chained =
exceptions. If some problem occurs, your application will have a chance =
to take care of it, instead of simply crashing. Also, the exception =
messages are more meaningful, giving a better picture of what went =
wrong.
Thanks to Claude Duguay for the ChainedException classes, and bug =
reports. And many thanks to the best tester of HTMLParser - Cedric Rosa =
- for countless bug reports - pls keep up the good work.
From the change log,=20
[1] Fixed bug 590250, problem in HTMLStringNode, by which a single
character on the last line was causing a parser crash
[2] Optimized and refactored HTMLParameterParaser.parseParameters()
[3] Modified PerformanceTest to exclude first reading in average =
computation
[4] Fixed bug in HTMLParameterParser.parseParameters(), due to which
params with spaces before =3D were not being picked up
[5] Made massive API changes - throwing exceptions and using =
HTMLEnumeration
[6] Fixed HTMLRemarkNode bug - we can recognize stuff like now.
[7] Fixed HTMLImageScanner bug - we can now fix image tags like
IMG SRC"somepic.jpg" - the missing equal to can be deduced
[8] Fixed HTMLLinkScanner bug - end tags within a link were not being =
included
inside the link data.
Please give your feedback regarding the API changes.=20
NOTE=20
[1] this release would break your existing applications due to the API =
change. Simply wrap the parsing in a try-catch block to cath a =
HTMLParserException and your apps should work again.
[2] There is one known bug (590703) caught by two testcases in =
parseParameters(). This is a minor bug which shouldnt affect =
applications, and should be fixed in the next release.
Cheers,
Somik
|
