W3C home > Mailing lists > Public > www-svg@w3.org > April 2008

SVGT 1.2 Tests: paint-stroke-205-t assumptions wrong?

From: Kalle Raita <kraita@nvidia.com>
Date: Fri, 25 Apr 2008 11:12:18 +0200
Message-ID: <85DD8D3F1695004A8A9A5AC16CE584BE06B836E3@deemmail01.nvidia.com>
To: <www-svg@w3.org>

Hi All,
I think that the test paint-stroke-205-t has some wrong assumptions. The
test is about
caps on zero length sub-paths. 

Implementation notes [1] says that the in the case of not finding a
length segment in the subpath, default tangent of (1, 0) is used.

In the test paint-stroke, the reference paths seem to assume (1, 1)
default tangent. 
I cannot find any rotation transforms in the test content that would
explain the 

As a specific example, let's take a segment starting at line 141 of the 
test case file.
<path d="M200,200L300,300M399.99,399.99L400.01,400.01M500,300L600,250"
  <title>Red comparison - if visible - error</title>
<path d="M200,200L300,300M400,400ZM500,300L600,250"
  <title>Blue testcase covers all red if correct</title>

The blue test path moves to (400,400) and immediately closes the
The subpath does not contain any segments with non-zero length, i.e., 
the default tangent is used. The red reference moves to (399.99,399.99) 
and draws line to (400.01, 400.01), generating a tangent with direction
(1,1), which contradicts the spec request to align "with the positive 
x-axis in user space". 
As a side note, the test case could be split over multiple files to make

it easier to map the SVG content to shapes on the screen.


   - Kalle Raita

Kalle Raita 
NVIDIA Corporation 
Tel. +358 40 723 1441 
http://eu.nvidia.com <http://eu.nvidia.com/>  

This email message is for the sole use of the intended recipient(s) and may contain
confidential information.  Any unauthorized review, use, disclosure or distribution
is prohibited.  If you are not the intended recipient, please contact the sender by
reply email and destroy all copies of the original message.
Received on Friday, 25 April 2008 09:13:55 UTC

This archive was generated by hypermail 2.3.1 : Wednesday, 8 March 2017 09:47:13 UTC