23082 - java class (jar files) for mozilla

Status: VERIFIED WONTFIX

(Core Graveyard :: Java: OJI, defect, P3)

Description

[Warren Harris]

What's the deal on shipping java classes with our release. Specifically the
following:

ifc11.jar -- ifc
jsj10.jar -- liveconnect
ldap10.jar -- ldap
nsplug50.jar -- scriptable (liveconnect-enabled) plugins (not supported now?)
nssec50.jar -- security stubs for backward compatibility with 4.x

These were included with mozilla classic, but not with the current mozilla. The
pre-build jar files used to be found in nav-java/stubs/classes, although I
don't remember where the real class files used to live.

I think we need a plan for whether we ship them or not.

Comment 1

[drapeau]

Agreed.  There was an email thread about this subject back in September (Chris
Hofmann might have a copy, and I *do* have a copy of some of that correspondence
that I am not sure if it's appropriate to copy here; don't know Netscape's
policy on putting that kind of information in public bug reports).  Anyway, the
gist of that conversation was "Hey, we should have marketing evaluate whether we
need to support these jar's, and coordinate with [e-mail]."  That
sounds good to me.  Seems like the key issue is whether any current Netscape
products or customers still rely on any of those jar's.  If forced to guess, I
would guess that the IFC stuff could be left out; I can't imagine anybody
actively writing Java apps is still using IFC.  On the other hand, LiveConnect
stuff seems like it should stay in, although I don't know the exact contents of
that jar.

By the way, another jar (iiop10.jar) was mentioned in the email correspondence
in September.

I'll raise this with Chris Saito, see if he can help us come up with a plan.  If
the answer is "Hey, George, you're the OJI module owner, it's really up to you
to decide," then I'll go ahead and make a decision, but it'll not likely be a
well-informed one.

My plan of attack would be to look into each of the jar's you mention in this
bug report and see if they're necessary for the operation of current Mozilla
functionality (e.g., jsj10.jar: is it necessary in order for LiveConnect to
work?).  If it's necessary, it stays; if not (e.g.,: IFC isn't necessary in
order for Mozilla/Comm5 to work), it goes.

Comment 2

[drapeau]

Not going to worry about this until a later Beta release, although I agree that 
it's important for the product.

Comment 3

[edburns]

George, what's the status on this?

Comment 4

[drapeau]

No change in status.  I'd call this an RFE for the next major release of the
commercial product, and Future it for Mozilla.  One person has asked for some
jars (Rob Weltman, iPlanet) for the commercial release, but didn't have any
supporting data about which customers would use it.  Nobody has asked, and it
seems late to be putting new stuff into the commercial product.

I'd mark this for much later.  It's a much less important bug at this point than
crasher bugs.

Comment 5

[Paul Wyskoczka]

Updating QA Contact

Comment 6

[edburns]

George, status?

Comment 7

[drapeau]

I will close this bug.  The Netscape 6 commercial product shipped without these
jar's, and they are not necessary for consumer browser support or Mozilla
support.  They would fall into the category of enterprise client support; if
eClient group implements the features that require these jars, they should
re-open the bug.

Until actual customers demand these jars, they should stay out of the download
package so as to reduce size and number of unnecesssary bits.

Comment 8

[John Unruh]

Verified per drapeau's comments.