18157 - Improperly treating nsresult as signed

Status: VERIFIED FIXED

(MailNews Core :: Backend, defect, P3)

Description

[Scott Furman]

As reported in n.p.m.builds by TenThumbs, an nsresult is being tested by
comparing less than zero, which will always fail since the nsresult type is
unsigned.  Need to use the NS_FAILED() macro instead.

http://lxr.mozilla.org/mozilla/source/mailnews/news/src/nsNNTPHost.cpp#359

Comment 1

[Scott Furman]

Similar compiler warnings also reported by TenThumbs:

http://lxr.mozilla.org/mozilla/source/mailnews/compose/src/nsMsgSend.cpp#1143
http://lxr.mozilla.org/mozilla/source/mailnews/compose/src/nsMsgSend.cpp#1146
http://lxr.mozilla.org/mozilla/source/mailnews/compose/src/nsMsgSend.cpp#1178
http://lxr.mozilla.org/mozilla/source/mailnews/compose/src/nsMsgSend.cpp#1181
http://lxr.mozilla.org/mozilla/source/mailnews/compose/src/nsMsgSend.cpp#1186
http://lxr.mozilla.org/mozilla/source/mailnews/compose/src/nsMsgSend.cpp#1197

Comment 2

[(not reading, please use [e-mail] instead)]

TenThumbs and fur, thanks for the heads up.

I've got the fix for the warning in nsNNTPHost.cpp in my tree.

re-assign to rhp so he can fix the errors in nsMsgSend.cpp.

Comment 3

[rhp (gone)]

Ok, I have these fixed in my tree and they are waiting to go into the tree when
it opens.

- rhp

Comment 4

[rhp (gone)]

My fix for the nsMsgSend.cpp is in place.

- rhp

Comment 5

[ppandit]

Checked nsINNTPHost.cpp and nsIMsgSend.cpp using LXR.

VERIFIED

Comment 6

[Pulsebot]

Pushed by ffxbld-merge:
https://hg.mozilla.org/releases/mozilla-beta/rev/cef0db10867c
[fenix] For https://github.com/mozilla-mobile/fenix/issues/18157: convert clock_ticks_per_second to quantity type.