536912 - SVG text node method getNumberOfChars() throws an exception.

Status: RESOLVED FIXED

(Core :: SVG, defect)

Description

[Luis Merino]

User-Agent:       Mozilla/5.0 (Macintosh; U; Intel Mac OS X 10_6_2; en-us) AppleWebKit/531.21.8 (KHTML, like Gecko) Version/4.0.4 Safari/531.21.10
Build Identifier: Mozilla/5.0 (Macintosh; U; Intel Mac OS X 10.6; en-US; rv:1.9.2b5) Gecko/20091204 Firefox/3.6b5

uncaught exception: [Exception... "Component returned failure code: 0x80004005 (NS_ERROR_FAILURE) [nsIDOMSVGTextPositioningElement.getNumberOfChars]" nsresult: "0x80004005 (NS_ERROR_FAILURE)" location: "JS frame :: http://localhost/assets/js/Raphael.class.js :: anonymous :: line 1435" data: no]

Reproducible: Always

Steps to Reproduce:
1. create a SVG text node
2. try to call its method "getNumberOfChars()"
3. get component NS_ERROR_FAILURE
Actual Results:  
uncaught exception: [Exception... "Component returned failure code: 0x80004005 (NS_ERROR_FAILURE) [nsIDOMSVGTextPositioningElement.getNumberOfChars]" nsresult: "0x80004005 (NS_ERROR_FAILURE)" location: "JS frame :: http://localhost/assets/js/Raphael.class.js :: anonymous :: line 1435" data: no]

Expected Results:  
to return a Javascript value native type "Number".

Comment 1

[Robert Longson [:longsonr]]

Please provide a complete minimal testcase. This functionality works for me but you can't call it on text that is display:none or that is not part of an SVG document.

Comment 2

[Luis Merino]

You were right on with this. If I call this method with an element that contains display="none" it throws this exception. In any case, why wasn't it failing in FF 3.5 and now in FF3.6 it's failing, is this a new feature?? just curious.

Comment 3

[Robert Longson [:longsonr]]

It's been like that ever since I wrote it. I've no idea what's going on in your case in 3.5.

Comment 4

[Dakota Ryan]

I'm also getting this error since upgrading to FF3.6. 

For a production test case, please visit http://carleton.aidcalculator.com/results/ (exception thrown and text nodes display incorrectly...) SVG text displays correctly in Webkit, IE7/8

This is has also been identified by Raphael JS users at: http://groups.google.com/group/raphaeljs/browse_thread/thread/8c3f94f7cf7b8c5b/2a4d54ece3f19370

Any thoughts on if FF3.6 broke this SVG method, or any leads on what may have changed to make the display stop functioning correctly in FF3.6?

Comment 5

[Robert Longson [:longsonr]]

Please attach a minimal testcase to the bug. I just get some kind of legalese page from http://carleton.aidcalculator.com/results/ and I don't know what to do with that :-(

Comment 6

[Dakota Ryan]

(In reply to comment #5)
> Please attach a minimal testcase to the bug. I just get some kind of legalese
> page from http://carleton.aidcalculator.com/results/ and I don't know what to
> do with that :-(

Correct. This site is in production, so there is legalese associated with the business logic. There is a javascript link at the bottom of the page that hides this legal text and reveals the SVG pie chart built with g.raphaeljs.com

Sorry for the confusion.

Comment 7

[Robert Longson [:longsonr]]

I'm unwilling to click though the licence because of its terms. Please provide a testcase as an attachment here.

Comment 8

[Robert Longson [:longsonr]]

For instance...

> User may not modify, copy, distribute, broadcast, transmit, reproduce, publish, mirror, frame, license, transfer, sell, or otherwise use any information or material obtained from or through this Estimator. 

I would need to do that.

> You may use the Estimator only as expressly permitted in these Terms of Use. In doing so, you must comply with any technical limitations in the Estimator that only allow you to use it in certain ways. You may not work around any technical limitations in the Estimator; reverse engineer, decompile or disassemble the Estimator

And reverse engineer at least part of it.

Comment 9

[Robert Longson [:longsonr]]

BTW if you are going to copy stuff from http://carleton.aidcalculator.com/results/ into bugzilla you had better be sure you have the legal right to do so and that we can use it to turn it into testcases.

Comment 10

[Grégory Teste]

Hello,
would that be ok for your tests : http://gregoo.ftp.free.fr/bugzilla.mozilla.org/536912/536912.xml ?

Comment 11

[Robert Longson [:longsonr]]

Indeed.

Comment 12

[Robert Longson [:longsonr]]

Attachment: patch

revert the logic to what it used to be.

Comment 13

[Robert Longson [:longsonr]]

pushed http://hg.mozilla.org/mozilla-central/rev/af0f0615944e

Comment 14

[Robert Longson [:longsonr]]

Comment on attachment 426276 [details] [diff] [review]
patch

This is a regression on 3.6 and affects raphael when used with SVG. The patch puts the logic back to the 3.5 logic and includes tests so should be very safe.

Comment 15

[Johnathan Nightingale [:johnath]]

Comment on attachment 426276 [details] [diff] [review]
patch

Didn't make 1.9.2.2, moving flag forward.

Comment 17

[Mike Beltzner [:beltzner, not reading bugmail]]

Comment on attachment 426276 [details] [diff] [review]
patch

a=beltzner for 1.9.2.4

Comment 18

[Robert Longson [:longsonr]]

pushed http://hg.mozilla.org/releases/mozilla-1.9.2/rev/183798d66e35

Comment 19

[Al Billings [:abillings - ex-MoCo]]

Verified fixed for 1.9.2 in Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.9.2.4pre) Gecko/20100412 Namoroka/3.6.4pre (.NET CLR 3.5.30729) using testcase in comment 10.