W3C home > Mailing lists > Public > public-css-testsuite@w3.org > January 2011

Re: Implementation report for WebToPDF.NET 1.0

From: Linss, Peter <peter.linss@hp.com>
Date: Fri, 21 Jan 2011 06:05:20 +0000
To: Robert Stam <robert@tallcomponents.com>
CC: "public-css-testsuite@w3.org" <public-css-testsuite@w3.org>
Message-ID: <0903A741-1FA4-4A1A-9A40-A73BBB36C12B@hp.com>
On Jan 20, 2011, at 5:28 AM, Robert Stam wrote:

> Hello All,
> 
> I have run the complete XHTML testsuite here with our WebToPDF.NET
> converter, which uses an implementation written completely in C#
> [Microsoft .NET] and doesn't use any open source software, it includes
> JavaScript and SVG as well. The only part we have to support is BIDI,
> which is reported as skipped in the attached implementation report.
> 
> More information can be found at:
> http://www.tallcomponents.com/webtopdf1.aspx
> 
> Please let us know if this file can be imported into the harness, so we
> can easily check our conformance compared with others.
> 
> 
> If you have any further questions, do not hesitate to contact me. 
> 
> 
> Best Regards,
> Robert Stam
> 
> TallComponents
> Follow us @ twitter: http://twitter.com/tallcomponents

Hi Robert, thanks for the data. 

I do have a concern however, your implementation report is also listing the reference files, which are not tests in themselves but must be compared to certain test files.

What's really bothering me is some of the results for the reference tests don't seem to make sense, for example:
xhtml1/block-in-inline-insert-015.xht	fail
xhtml1/block-in-inline-insert-015-nosplit-ref.xht	pass
xhtml1/block-in-inline-insert-015-ref.xht	pass

and:
xhtml1/text-indent-wrap-001.xht	pass
xhtml1/text-indent-wrap-001-notref-block-margin.xht	pass

which should not be providing matching output... so how are you determining that both 'pass'?

This makes me question the validity of the references tests in the report... can you please clarify?


Peter

(You're also reporting results for the toc and chapter files, which are not part of the test suite...)
((On the positive side, using your results, 11 of our previously blocked tests now have two passing implementations))
Received on Friday, 21 January 2011 06:08:20 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Friday, 21 January 2011 06:08:27 GMT