I strongly disagree. To argue the flip side of your point.......If we go with
the .jar extension, then it makes it difficult to associate webapps with
specialized tools (and there *will* be several of them) for manipulating/viewing
them. Once a decent tool comes about to manipulate webapps, how will the OS be
able to determine that the webapp tool is needed to manipulate the file? Now
that I have JDK 1.2 on my machine, my assocations for JAR files have have stolen
by the JDK to support the new executable jar mechanisms. I prefer to have the a
special extension so that I can kick off the specized tool. To go a step
further, I would prefer to have ejb jar files get their own extension as well so
that I can kick off specialized tools for them as well. You can always
associate multiple types with the same tool (Winzip=*.jar,*.zip,*.war for
example), but it's much more difficult to associate the same extension with
multiple tools.
The alternative is to be able to associate JARs with some extensible tool (that
would ship with the JDK) that can be extended to kick off other tools based on
the contents of the jar. Since this doesn't exist today, and there won't be a
JDK with new tools for quite some time, I don't think this is an option.
I say give webapp jars a special extension!
-----
Spike Washburn
IBM WebSphere Application Server
Internet E-mail: [EMAIL PROTECTED]
Frank Carver <[EMAIL PROTECTED]> on 07/06/99 05:05:03 AM
Please respond to "A mailing list for discussion about Sun Microsystem's Java
Servlet API Technology." <[EMAIL PROTECTED]>
To: [EMAIL PROTECTED]
cc: (bcc: Donald Washburn/Raleigh/IBM)
Subject: Re: Comments on 2.2 Public Review Draft
On Tuesday, July 06, 1999 7:15 AM, Ted Neward [SMTP:[EMAIL PROTECTED]] wrote:
> Craig McClanahan wrote:
> > A "Web Application Archive" file is in fact a standard JAR file, with
the
> > extra configuration stuff in a /WEB-INF subdirectory.
> >
> Then why not simply mandate the .jar file must have a /WEB-INF
subdirectory
> in order to be treated as a Web Application Archive? Why create a new
> extension?
I strongly agree. As I tried to point out in my comments, adding a new
"file type"
to a system is a bad thing. It means that work may have to be done on every
client system (millions) and many server systems to do some or all of the
following:
associate the file extension with a mime type
define a valid set of operations on these files
define a default operation on these files
set up a visual representation for these files
And will also probably need changes to many applications which need to
understand
and process these files (eg. WinZip etc.)
This seems completely mad to me, when by simply adding another optional
section
to the META-INF structure of the existing jar file spec, all of this could
be avoided.
The only counter-argument I can imagine is that some software may want/need
to
do different things to "jar" and "war" files, _based_solely_on_extension_.
I can't think
of a real, unavoidable, example of this usage though. Can anyone?
Frank.
--
Frank Carver
[ Personal: [EMAIL PROTECTED] http://www.efsol.com/ ]
[ At Work: [EMAIL PROTECTED] tel +44 (0)1473 227371 ]
___________________________________________________________________________
To unsubscribe, send email to [EMAIL PROTECTED] and include in the body
of the message "signoff SERVLET-INTEREST".
Archives: http://archives.java.sun.com/archives/servlet-interest.html
Resources: http://java.sun.com/products/servlet/external-resources.html
LISTSERV Help: http://www.lsoft.com/manuals/user/user.html
___________________________________________________________________________
To unsubscribe, send email to [EMAIL PROTECTED] and include in the body
of the message "signoff SERVLET-INTEREST".
Archives: http://archives.java.sun.com/archives/servlet-interest.html
Resources: http://java.sun.com/products/servlet/external-resources.html
LISTSERV Help: http://www.lsoft.com/manuals/user/user.html
