1300007 - Don't close a stream in an assertion

Status: RESOLVED FIXED

(Core :: DOM: Core & HTML, defect)

Description

[Nicholas Nethercote [inactive]]

Because assertions disappear in non-debug builds.

Comment 1

[Nicholas Nethercote [inactive]]

Attachment: Don't close a stream in an assertion

Comment 2

[Nicholas Nethercote [inactive]]

Attachment: Don't close a stream in an assertion

New version that (a) compiles in opt builds, and (b) doesn't depend on another
patch in my queue, which means it will backport more easily.

Comment 3

[Nicholas Nethercote [inactive]]

Comment on attachment 8787519 [details] [diff] [review]
Don't close a stream in an assertion

Review of attachment 8787519 [details] [diff] [review]:
-----------------------------------------------------------------

Carrying over r+.

Comment 4

[Nicholas Nethercote [inactive]]

https://hg.mozilla.org/integration/mozilla-inbound/rev/d6246f5ede4a75deda1b0294be758cb4aad855de
Bug 1300007 - Don't close a stream in an assertion. r=baku.

Comment 5

[Pulsebot]

Pushed by [e-mail]:
https://hg.mozilla.org/integration/mozilla-inbound/rev/d6246f5ede4a
Don't close a stream in an assertion. r=baku.

Comment 6

[Wes Kocher (:KWierso) (Not reading bugmail; email directly if needed)]

https://hg.mozilla.org/mozilla-central/rev/d6246f5ede4a

Comment 7

[Nicholas Nethercote [inactive]]

Comment on attachment 8787519 [details] [diff] [review]
Don't close a stream in an assertion

Approval Request Comment

[Feature/regressing bug #]: bug 994190 introduced the bug.

[User impact if declined]: Files failing to be closed when they should, causing file descriptors to be leaked. It might also cause memory leaks in some cases, but I'm not certain about that.

[Describe test coverage new/current, TreeHerder]: the line of code in question runs in debug builds. It is unintentionally excluded in non-debug builds.

[Risks and why]: Extremely low. It affects a single function call, and only changes behaviour in non-debug builds.

[String/UUID change made/needed]: none.

Comment 8

[Sylvestre Ledru [:Sylvestre]]

If I understand correctly, this is clearly not a recent regression. As we are super late in the 49 cycle, I would rather see that rides the train with 50

Comment 9

[Carsten Book [:Tomcat]]

https://hg.mozilla.org/releases/mozilla-aurora/rev/efc8bd58d7c8