288265 - save page completely is copying id to next specification value

Status: RESOLVED FIXED

(Core :: SVG, defect)

Description

[Heinz Kohl]

User-Agent:       Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.8b2) Gecko/20050329
Build Identifier: Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.8b2) Gecko/20050329

saving a complete page using Mozilla SVG is producing values like:

<svg:marker id="MdartA" viewBox="MdartA0 0 20 20" refX="0" refY="10"
orient="auto" markerWidth="10" markerHeight="10">

 from source:
<svg:marker id="MdartA" viewBox="MdartA0 0 20 20" refX="0" refY="10"
orient="auto" markerWidth="10" markerHeight="10">

copying the id to the beginning of the next specification value

Reproducible: Always

Steps to Reproduce:
1.load the appended xml-file 
2.save page complete
3.view page will show the result

Actual Results:  
finding this result was due to the surprising marker size after saving the file
and correcting the doctype (maybe related error, Bug# 288133


I've not tested non-SVG browser reaction (sorry, no time at the moment, has to
report another maybe related bug and then to go for my normal work)

Comment 1

[Heinz Kohl]

Attachment: xml-file showing the described error situation

just showing the described error situation

Comment 2

[Jonathan Watt [:jwatt]]

I presume that you mean the source was:

<svg:marker id="MdartA" viewBox="0 0 20 20" refX="0" refY="10"
orient="auto" markerWidth="10" markerHeight="10">

I can't reproduce this using the file you attached. Note I had to remove the
'id' attribute values that had been carried to the next attribute since you
attached the file with the errors in it. To help in reproducing this can you
describe what you did in detail. Did you save the file from a website? Your HDD?
Does it matter if you correct the doctype before you try to save it?

Comment 3

[Heinz Kohl]

Attachment: the correct source file, not the result of saving

sorry, I appended the file after saving ...

Comment 4

[tor]

Attachment: nsSVGRect::GetValue was appending instead of replacing

Comment 5

[Asa Dotzler [:asa]]

Has this been tested in the non-svg case to see that it doesn't break anything?
what's the risk here and is it worth taking for alpha if we're going to try to
ship rsn?

Comment 6

[tor]

This is a SVG-only change.

Comment 7

[Daniel Veditz [:dveditz]]

Comment on attachment 183093 [details] [diff] [review]
nsSVGRect::GetValue was appending instead of replacing

a=dveditz

Comment 8

[tor]

Checked in.