530032 - xpcshell-tests in /mailnews/db/ fail on JS strict mode

Status: RESOLVED FIXED

(MailNews Core :: Backend, defect)

Description

[Jae-Seong Lee-Russo]

User-Agent:       Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.9.3a1pre) Gecko/20091119 Minefield/3.7a1pre (.NET CLR 3.5.30729)
Build Identifier: 

xpcshell uni tests in /mailnews/db/ fail on JS strict mode

Reproducible: Always

Comment 1

[Jae-Seong Lee-Russo]

Attachment: Fix JS strict warnings

Comment 2

[David :Bienvenu]

Comment on attachment 413560 [details] [diff] [review]
Fix JS strict warnings

thx for looking at this.

this looks like more than a warning fix, and you'll need to ask :asuth about it:

+  contactLikeQuery.nameLike(contactLikeQuery.WILDCARD, personNameSubstring,
+                            contactLikeQuery.WILDCARD);

+var gMessageGenerator;
+

since this is only used in setupGlobals, it can be changed to 

let gMessageGenerator = new MessageGenerator();

and in fact, since it's not a global, the name could just be messageGenerator.

Comment 3

[Jae-Seong Lee-Russo]

Attachment: Address comment #2

reference to undefined property contactLikeQuery.WILD

Comment 4

[Jae-Seong Lee-Russo]

Comment on attachment 413756 [details] [diff] [review]
Address comment #2

> You don't need sr for test-only patches.

Comment 5

[Mark Banner (:standard8)]

Checked in: http://hg.mozilla.org/comm-central/rev/29b8e3f6c26e

Sorry, I forgot to set the correct username on push.

Comment 6

[Serge Gautherie (:sgautherie)]

Comment on attachment 413756 [details] [diff] [review]
Address comment #2

I guess this (fix) would be wanted on c-1.9.1 too.
Should be no risk.

Comment 7

[Mark Banner (:standard8)]

(In reply to comment #3)
> Created an attachment (id=413756) [details]

Andrew, am I right that without the mailnews/ changes in this patch we'll function exactly the same (with the addition of a strict warning)? i.e. the only bug this patch fixes is the strict warning?

Comment 8

[Andrew Sutherland [:asuth] (he/him)]

(In reply to comment #7)
> Andrew, am I right that without the mailnews/ changes in this patch we'll
> function exactly the same (with the addition of a strict warning)? i.e. the
> only bug this patch fixes is the strict warning?

Yes, I believe the WILD/WILDCARD fix does not actually affect the code flow.  I think in a previous investigation I noticed the problem and got concerned then noticed the control flow ends up in the same place no matter what and forgot about the problem.

Comment 9

[Mark Banner (:standard8)]

Comment on attachment 413756 [details] [diff] [review]
Address comment #2

Based on Andrew's comment, I don't think we need this on branch.