770598 - Clean up NS_NewSVGDocument()

Status: RESOLVED FIXED

(Core :: SVG, defect)

Description

[Daniel Holbert [:dholbert]]

decoder was just asking me about some unused-by-tests code in nsSVGDocument.cpp, and one of the lines for checking the result of "new" in  NS_NewSVGDocument().

I'm filing this bug on removing that and doing a bit of other fixup in that function (using a nsRefPtr rather than a manual ADDREF / RELEASE)

Comment 1

[Daniel Holbert [:dholbert]]

Attachment: fix

This matches the pattern that we use in NS_NewSVG##_elementName##Element(), with nsRefPtr etc.:
http://mxr.mozilla.org/mozilla-central/source/content/svg/content/src/nsSVGElement.h?mark=584-602#581

The only difference is that I removed the null-check after "new", since it's infallible.

Comment 2

[Jonathan Watt [:jwatt]]

Comment on attachment 638810 [details] [diff] [review]
fix

longsonr is on vacation for a few weeks, so taking review.

Comment 3

[Daniel Holbert [:dholbert]]

Thanks!
https://hg.mozilla.org/integration/mozilla-inbound/rev/db774ca70b42

Comment 4

[:Ms2ger (he/him; ⌚ UTC+1/+2)]

Comment on attachment 638810 [details] [diff] [review]
fix

Review of attachment 638810 [details] [diff] [review]:
-----------------------------------------------------------------

::: content/svg/document/src/nsSVGDocument.cpp
@@ +112,4 @@
>      return rv;
>    }
>  
> +  *aInstancePtrResult = doc.forget().get();

Exactly what doc.forget(aInstancePtrResult) is for

Comment 5

[Daniel Holbert [:dholbert]]

Ah, nice -- I thought there was something like that.  (FWIW, I was copypasting an existing pattern, for maximum foolproof-ness, as noted in comment 1.)

Probably worth doing an MXR search for "forget().get()" and replacing with forget().  Ms2ger, perhaps you'd be interested in filing and either fixing or mentoring? :)

Comment 6

[Ryan VanderMeulen [:RyanVM]]

https://hg.mozilla.org/mozilla-central/rev/db774ca70b42