141317 - CopyToLowerCase produces strings with random content

Status: VERIFIED FIXED

(Core :: XPCOM, defect)

Description

[Kai Engert (:KaiE:)]

CopyToLowerCase produces strings with random content

It only copies the chars that need conversion.
More related helper classes seem to have this bug.
I'll attach a fix.

Comment 1

[Kai Engert (:KaiE:)]

Attachment: Suggested fix

This fixes 4 methods, all having the same bug.

Comment 2

[Kai Engert (:KaiE:)]

Attachment: Updated fix

Jag convinced me that only 2 of the 4 locations require my change, because the
convert classes already have the buffer filled.

Comment 3

[Kai Engert (:KaiE:)]

Jag, Alec, can you please r/sr ?

Comment 4

[Alec Flett]

Comment on attachment 81774 [details] [diff] [review]
Updated fix

bwah!
good catch!
sr=alecf

Comment 5

[jag (Peter Annema)]

Comment on attachment 81774 [details] [diff] [review]
Updated fix

r=jag

Comment 6

[Kai Engert (:KaiE:)]

Jag, do we want this patch on the branch? Nominating.

Comment 7

[Kai Engert (:KaiE:)]

Checked in to trunk, fixed.

Comment 8

[Kai Engert (:KaiE:)]

Actually, I need to nominate this bug for branch checkin, because it blocks bug
130692.

Comment 9

[jag (Peter Annema)]

Yes, I suggest we take this on the branch.

Comment 10

[Scott Collins]

Comment on attachment 81774 [details] [diff] [review]
Updated fix

a=scc for checkin to the mozilla1.0 branch

Comment 11

[Jaime Rodriguez, Jr.]

let's get this one in after rc2. adt1.0.0- [adt2 rtm]

Comment 12

[Judson Valeski]

this is a nasty bug. how did we come across it though? is it botching our UI
somewhere?

Comment 13

[Kai Engert (:KaiE:)]

Jud, I came across it when I produced a patch for bug 130692.

Comment 14

[Kai Engert (:KaiE:)]

Re-nominating.

Comment 15

[Jaime Rodriguez, Jr.]

Would [e-mail] be the right person to verify this one on the trunk?

Comment 16

[Randell Jesup [:jesup] (needinfo me)]

a=[e-mail] re-approved for branch checkin

Comment 17

[Kai Engert (:KaiE:)]

Jaime, this bug can not yet be verified by QA, it can only be verified by
looking at the code.

However, there are several areas in Mozilla that use this piece of code, and it
could potentially cause random problems, that might be difficult to reproduce.
We just don't know yet a reproducable effect of this bug.

Look at comment 4 and comment 12 for reactions about the seriousness of this bug.

I recommend to check this code in to the branch. It was checked in to the trunk
two weeks ago and nobody complained since then.

Comment 18

[Jaime Rodriguez, Jr.]

thanks for the explanation Kaie.

adt1.0.0+ (on ADT's behalf) for approval to checkin to the 1.0 branch.  After,
checking in, please add the fixed1.0 keyword.

Comment 19

[Kai Engert (:KaiE:)]

Checked in to branch.

Comment 20

[John Unruh]

Verified per kaie's comment.