376395 - New version of JEP (0.9.6.1), please land on trunk and branches

Status: VERIFIED FIXED

(Core Graveyard :: Plug-ins, defect)

Description

[Steven Michaud [:smichaud] (Retired)]

Attachment: Change log for JEP 0.9.6.1

This version contains a bunch of bugfixes.  No one of them is
earthshaking, but together I think they make this a significant
update.

Please land JEP 0.9.6.1 on the trunk and branches as soon as
conveniently possible.  I'd like to get it into Firefox 1.5.0.12 and
2.0.0.4.

http://javaplugin.sourceforge.net/

Those who want to try the new version right away will need to install
it "over" the older versions currently bundled with Mozilla.org
browsers.  I recommend installing the new JEP to your
/Library/Internet Plug-Ins/ folder, then removing older copy(ies) of
the JEP from your Mozilla.org browser(s).  For more information see
the JEP Readme.

Comment 1

[Josh Aas]

Comment on attachment 260496 [details]
Change log for JEP 0.9.6.1

Looks good.

Comment 2

[Mike Pinkerton (not reading bugmail)]

Comment on attachment 260496 [details]
Change log for JEP 0.9.6.1

rs=pink

Comment 3

[Josh Aas]

landed on trunk

Comment 4

[Marcia Knous [:marcia]]

Looks good on today's Mac build on my PPC running 10.3.9, Mozilla/5.0 (Macintosh; U; PPC Mac OS X Mach-O; en-US; rv:1.9a4pre) Gecko/2007040504 Minefield/3.0a4pre. On my Intel Mac running the same build it still shows the 0.9.6 version.

Comment 5

[Marcia Knous [:marcia]]

Correction - it looks fine with a fresh profile. My existing crufty Intel profile showed the old version.

(In reply to comment #4)
> Looks good on today's Mac build on my PPC running 10.3.9, Mozilla/5.0
> (Macintosh; U; PPC Mac OS X Mach-O; en-US; rv:1.9a4pre) Gecko/2007040504
> Minefield/3.0a4pre. On my Intel Mac running the same build it still shows the
> 0.9.6 version.
> 

Comment 6

[Daniel Veditz [:dveditz]]

How safe and well-tested is this JEP release? That's a fair number of fixes for problems that as you say are not earthshaking. The gain, in other words, looks relatively small so we worry about the risk of making things less stable. (on the other hand some of those rare crashes might not be happy crashes.)

Not turning this down, looking for reassurance from a quality/stability standpoint.

Comment 7

[Ray Kiddy]

I have lived on it at various times with various releases and have not seen problems. That being said, I do not know of a "test plan" that one can point to.

Given ad hoc testing, how can rational decisions about risk be made? Without tracking of when tests get executed, how can one judge what has been done? Without specifications, how can testing be anything other than ad hoc? It is a challenge.

What kinds of tests would address the concerns you have? Capturing that now would be _extremely_ valuable.

Comment 8

[Steven Michaud [:smichaud] (Retired)]

(In reply to comment #6)

I've tested it very thoroughly (as I've done for all JEP releases for
at least the last year).  Each new version fixes things, of course.
But I also try to be as careful as possible not to break things in my
new releases.  Sometimes I miss something ... but my record is better
than 95% (at least) of other programs :-)

Even if that weren't true, though, I think the JEP should be landed on
the branches as soon as possible.  The longer it's on the branches
before a new branch release, the more it will get tested, and the more
likely that any serious problem (one I missed) will get found out.

Landing it on the trunk doesn't really get the JEP the kind of testing
I want -- the trunk is almost always full of other problems, which
tend to drown out anything that might be wrong with the JEP.

Comment 9

[Josh Aas]

I can vouch for JEP's record here. I think it should go on the branch. I have also tested this particular version of JEP myself, though probably not as thoroughly as Steven has.

Comment 10

[Steven Michaud [:smichaud] (Retired)]

My own testing procedure is to keep going through a list of
"difficult" sites that I've build up over time, on each of the
different major versions of Mac OS X, with each of the different Java
versions (e.g. 1.4.2, 5.0 and 6.0).  I'll usually repeat my tests
after every non-trivial change to the JEP source code.

(My list of sites is on record at bug 371084 comment #3.)

Yes, I suppose this is an ad hoc procedure.  But I think it's worked
remarkably well.  One sign of this (which is apparent from my change
logs) is that I almost always discover (and fix) more bugs than
everyone else put together.

I slowly keep adding new sites.  But I make sure that each one is in
some way "challenging" -- otherwise testing with it would just be
wasting my time.

Comment 11

[Marcia Knous [:marcia]]

Steven: If you would kindly share those test cases, I can add them to Litmus and we can have a JEP regression suite that can be run by QA and the community. I am also willing to run trunk and branch and give this a thorough workout.

(In reply to comment #10)
> My own testing procedure is to keep going through a list of
> "difficult" sites that I've build up over time, on each of the
> different major versions of Mac OS X, with each of the different Java
> versions (e.g. 1.4.2, 5.0 and 6.0).  I'll usually repeat my tests
> after every non-trivial change to the JEP source code.
> 
> (My list of sites is on record at bug 371084 comment #3.)
> 
> Yes, I suppose this is an ad hoc procedure.  But I think it's worked
> remarkably well.  One sign of this (which is apparent from my change
> logs) is that I almost always discover (and fix) more bugs than
> everyone else put together.
> 
> I slowly keep adding new sites.  But I make sure that each one is in
> some way "challenging" -- otherwise testing with it would just be
> wasting my time.
> 

Comment 12

[Steven Michaud [:smichaud] (Retired)]

> If you would kindly share those test cases

See bug 371084 comment #3 :-)

And I list another site in bug 371084 comment #4.

Comment 13

[Steven Michaud [:smichaud] (Retired)]

(Following up comment #12)

But also note what I say in bug 371084 comment #6 -- three of the
sites are internet banking sites, and if too many people test them at
once they'll think they're under attack.

Comment 14

[Daniel Veditz [:dveditz]]

Comment on attachment 260496 [details]
Change log for JEP 0.9.6.1

approved for 1.8.0.12 and 1.8.1.4, a=dveditz for release-drivers

Comment 15

[Josh Aas]

Checked in on MOZILLA_1_8_BRANCH and MOZILLA_1_8_0_BRANCH

Comment 16

[Marcia Knous [:marcia]]

verified on the 1.8 branch using Mozilla/5.0 (Macintosh; U; PPC Mac OS X Mach-O; en-US; rv:1.8.1.4pre) Gecko/2007041103 BonEcho/2.0.0.4pre. I was using my PPC Mac running 10.3.9 with a new profile. Will verify on an Intel box as well. I will also verify on the 1.8.0 branch shortly.

Comment 17

[Marcia Knous [:marcia]]

verified on the 1.8.0 branch using Mozilla/5.0 (Macintosh; U; PPC Mac OS X Mach-O; en-US; rv:1.8.0.12pre) Gecko/20070411 Firefox/1.5.0.12pre. Running on a PPC Mac (10.3.9) with a new profile. Will verify on an Intel box shortly.

Comment 18

[Chris Lawson (gone)]

I'm fairly certain this update broke access to Second Stage SABRE (either via the traditional Java front-end or the Java-based Citrix front-end) for American Airlines and American Eagle employees. More information coming when I can do some further testing this afternoon.

Comment 19

[Chris Lawson (gone)]

Sorry for the false alarm; not sure what the problem was but I seem to have solved it.

Verified on trunk using Camino 2007040902 (1.2+).

cl

Comment 20

[Steven Michaud [:smichaud] (Retired)]

I've just released JEP 0.9.6.2, which contains a fix for bug 377215 (a
regression that's been designated a blocker for 1.8.1.4 and 1.8.0.12).
For more information see that bug and bug 377779.

Comment 21

[Samuel Sidler (old account; do not CC)]

Marking this bug as verified as well, although it's superceded by bug 377779.