Changing length of a line also changes line width (transform)
Affects | Status | Importance | Assigned to | Milestone | |
---|---|---|---|---|---|
Inkscape |
Fix Released
|
Medium
|
Alvin Penner |
Bug Description
Inkscape 0.48+devel r12832
Tested on Windows XP and Windows 7, same behaviour.
First of all: This behaviour is not a result of having activated "when scaling objects, scale the stroke width by same proportion". It is disabled.
Steps to reproduce:
* Preferences --> Behavior --> Transform --> Store transformation = Preserved
* Draw a straigt horisontal line with two nodes. Make it long.
* Activate selec/move tool, and select the line.
* Duplicate the line and place it close to the first one (so that any changes in line width will be easy to see).
* Use the select/move tool to make the duplicated line short
* Expected behavior is that the short line have the same width as the original line, but it have being wider. Also the stroke style --> Width property of the short line now doesn't match the width as it appear on schreen.
Last stable version (0.48.4) is not affected. A file created by r12832 where line thickness is displayed wrong will appear correct when opened in stable version of Inkscape.
Changed in inkscape: | |
status: | New → Incomplete |
Changed in inkscape: | |
assignee: | nobody → Alvin Penner (apenner) |
importance: | Undecided → Medium |
milestone: | none → 0.91 |
tags: | added: transformations |
Changed in inkscape: | |
status: | Fix Committed → Fix Released |
> * Preferences --> Behavior --> Transform --> Store transformation = Preserved
> (…)
> * Expected behavior is that the short line have the same width as the original line
Your expectation is wrong - Inkscape's new renderer renders the stroke width of the path with a preserved transformatio correctly in accordance with the SVG 1.1 specification (the older renderer in stable is wrong, bug #165715). Just view it with a current web browser …
Any special reason you change the preferences to keep transformations preserved (this is not the default setting)?
<off-topic>
Current trunk (tested with r12934) has a regression when reproducing the reported issue, but AFAICT at first sight that regression is not related to what pRototype reports here.
</off-topic>