Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]


Groups > comp.lang.java.programmer > #29980 > unrolled thread

Grief!

Started byKnute Johnson <nospam@rabbitbrush.frazmtn.com>
First post2014-09-14 18:39 -0700
Last post2014-09-15 14:13 -0700
Articles 9 — 7 participants

Back to article view | Back to comp.lang.java.programmer


Contents

  Grief! Knute Johnson <nospam@rabbitbrush.frazmtn.com> - 2014-09-14 18:39 -0700
    Re: Grief! Arne Vajhøj <arne@vajhoej.dk> - 2014-09-14 21:51 -0400
      Re: Grief! Knute Johnson <eternal@knutejohnson.com> - 2014-09-15 14:09 -0700
        Re: Grief! Eric Sosman <esosman@comcast-dot-net.invalid> - 2014-09-15 17:32 -0400
          Re: Grief! Knute Johnson <eternal@knutejohnson.com> - 2014-09-15 15:49 -0700
    Re: Grief! Wayne <nospam@all.invalid> - 2014-09-14 22:14 -0400
    Re: Grief! Kevin McMurtrie <mcmurtrie@pixelmemory.us> - 2014-09-14 21:23 -0700
      Re: Grief! Josip Almasi <joe@vrspace.org> - 2014-09-15 17:25 +0200
      Re: Grief! Knute Johnson <eternal@knutejohnson.com> - 2014-09-15 14:13 -0700

#29980 — Grief!

FromKnute Johnson <nospam@rabbitbrush.frazmtn.com>
Date2014-09-14 18:39 -0700
SubjectGrief!
Message-ID<lv5fvs$12r$1@dont-email.me>
Two days of grief, my manifest file with only one line in it:

Permissions:         sandbox

Java kept rejecting the app as insecure.

knute...

[toc] | [next] | [standalone]


#29981

FromArne Vajhøj <arne@vajhoej.dk>
Date2014-09-14 21:51 -0400
Message-ID<5416463d$0$299$14726298@news.sunsite.dk>
In reply to#29980
On 9/14/2014 9:39 PM, Knute Johnson wrote:
> Two days of grief, my manifest file with only one line in it:
>
> Permissions:         sandbox
>
> Java kept rejecting the app as insecure.

Unsigned Applet and newer Java versions?

Arne

[toc] | [prev] | [next] | [standalone]


#29990

FromKnute Johnson <eternal@knutejohnson.com>
Date2014-09-15 14:09 -0700
Message-ID<lv7khu$cs7$1@dont-email.me>
In reply to#29981
On 9/14/2014 18:51, Arne Vajhøj wrote:
> On 9/14/2014 9:39 PM, Knute Johnson wrote:
>> Two days of grief, my manifest file with only one line in it:
>>
>> Permissions:         sandbox
>>
>> Java kept rejecting the app as insecure.
>
> Unsigned Applet and newer Java versions?
>
> Arne

It's a signed Java Web Start app that runs in the sandbox.  I have a 
code signing cert and the file is signed.  The problem was the blank 
spaces between 'Permissions:' and 'sandbox', only one space is 
permissible between the colon and the key or value.  I couldn't find 
anything in the docs but that is the way it works.  I've never seen a 
properties file or any other pair file that wouldn't ignore blank space.

Took two days to find that.  I'm officially insane now.

-- 

Knute Johnson

[toc] | [prev] | [next] | [standalone]


#29992

FromEric Sosman <esosman@comcast-dot-net.invalid>
Date2014-09-15 17:32 -0400
Message-ID<lv7ltk$ld3$1@dont-email.me>
In reply to#29990
On 9/15/2014 5:09 PM, Knute Johnson wrote:
> On 9/14/2014 18:51, Arne Vajhøj wrote:
>> On 9/14/2014 9:39 PM, Knute Johnson wrote:
>>> Two days of grief, my manifest file with only one line in it:
>>>
>>> Permissions:         sandbox
>>>
>>> Java kept rejecting the app as insecure.
>>
>> Unsigned Applet and newer Java versions?
>>
>> Arne
>
> It's a signed Java Web Start app that runs in the sandbox.  I have a
> code signing cert and the file is signed.  The problem was the blank
> spaces between 'Permissions:' and 'sandbox', only one space is
> permissible between the colon and the key or value.  I couldn't find
> anything in the docs but that is the way it works.  I've never seen a
> properties file or any other pair file that wouldn't ignore blank space.

     Only one space?  Sounds moronic.[*]  After some digging, though, I
found these syntax productions in Oracle's "JAR File Specification" at
<http://docs.oracle.com/javase/7/docs/technotes/guides/jar/jar.html#JAR_Manifest>:

	header:  name : value
	...
	value:   SPACE *otherchar newline *continuation

So, the single SPACE character is actually a documented requirement of
the syntax, odd[*] as that may seem.

     [*] Maybe not completely moronic nor totally odd, though.  The
syntax has no quoting convention, so if the parser were to collapse a
run of spaces together there'd be no way to specify a space as the
first character of a value -- and what a huge loss THAT would be!

-- 
esosman@comcast-dot-net.invalid

[toc] | [prev] | [next] | [standalone]


#29993

FromKnute Johnson <eternal@knutejohnson.com>
Date2014-09-15 15:49 -0700
Message-ID<lv7qdc$jnk$1@dont-email.me>
In reply to#29992
On 9/15/2014 14:32, Eric Sosman wrote:
> On 9/15/2014 5:09 PM, Knute Johnson wrote:
>> On 9/14/2014 18:51, Arne Vajhøj wrote:
>>> On 9/14/2014 9:39 PM, Knute Johnson wrote:
>>>> Two days of grief, my manifest file with only one line in it:
>>>>
>>>> Permissions:         sandbox
>>>>
>>>> Java kept rejecting the app as insecure.
>>>
>>> Unsigned Applet and newer Java versions?
>>>
>>> Arne
>>
>> It's a signed Java Web Start app that runs in the sandbox.  I have a
>> code signing cert and the file is signed.  The problem was the blank
>> spaces between 'Permissions:' and 'sandbox', only one space is
>> permissible between the colon and the key or value.  I couldn't find
>> anything in the docs but that is the way it works.  I've never seen a
>> properties file or any other pair file that wouldn't ignore blank space.
>
>      Only one space?  Sounds moronic.[*]  After some digging, though, I
> found these syntax productions in Oracle's "JAR File Specification" at
> <http://docs.oracle.com/javase/7/docs/technotes/guides/jar/jar.html#JAR_Manifest>:
>
>
>      header:  name : value
>      ...
>      value:   SPACE *otherchar newline *continuation
>
> So, the single SPACE character is actually a documented requirement of
> the syntax, odd[*] as that may seem.
>
>      [*] Maybe not completely moronic nor totally odd, though.  The
> syntax has no quoting convention, so if the parser were to collapse a
> run of spaces together there'd be no way to specify a space as the
> first character of a value -- and what a huge loss THAT would be!
>

Exactly!

-- 

Knute Johnson

[toc] | [prev] | [next] | [standalone]


#29982

FromWayne <nospam@all.invalid>
Date2014-09-14 22:14 -0400
Message-ID<KXrRv.263063$O13.218430@fx17.iad>
In reply to#29980
On 9/14/2014 9:39 PM, Knute Johnson wrote:
> Two days of grief, my manifest file with only one line in it:
> 
> Permissions:         sandbox
> 
> Java kept rejecting the app as insecure.
> 
> knute...

I had similar pain, and did some research on it.  Here's
what I learned:

Starting with Java 7U51, the user can add exceptions to
the rule that blocks unsigned or self-signed apps from
running, by white-listing URLs using the Java control
panel.  The exception list is a list of sites from which
you will allow unsigned or self-signed code to run.  You
should add a couple of entries to enable running code
from your local computer:

http://localhost/
http://127.0.0.1/
https://localhost/
https://127.0.0.1/
file:///

Only the file:/// entry is needed unless you're running a
local server such as a web server, Tomcat, Glassfish, etc.,
and plan on serving up Java applets and/or WebStart
applications.  In that case, you may need to specify a
non-standard port number.  For example, by default the
Glassfish server uses port 8080, so you need to add
“http://localhost:8080/” and “https://localhost:8080/”.

Note that IPv6 isn't currently supported by this exception
list; also the JRE doesn't resolve name to IP addresses, so
you need both the host name as well has the IP addresses
added.  Also note you will get a warning with http: and
file: protocols.  Since you are using localhost, it doesn't
matter so you should ignore those warnings.

For full details, you may which to view the web page I put
together on code signing (which includes information on
the required manifest entries):

   <http://wpollock.com/AJava/SignedCodeDemo/>

-- 
Wayne

[toc] | [prev] | [next] | [standalone]


#29985

FromKevin McMurtrie <mcmurtrie@pixelmemory.us>
Date2014-09-14 21:23 -0700
Message-ID<mcmurtrie-380B67.21230014092014@news.sonic.net>
In reply to#29980
In article <lv5fvs$12r$1@dont-email.me>,
 Knute Johnson <nospam@rabbitbrush.frazmtn.com> wrote:

> Two days of grief, my manifest file with only one line in it:
> 
> Permissions:         sandbox
> 
> Java kept rejecting the app as insecure.
> 
> knute...

From what I've read in the Oracle bug database, Applet permissions 
essentially don't work any more.  I found this out by needing to get an 
Applet running - only one Java version works despite the Applet being 
correctly built and deployed.

I'd give up on running Java in a browser.  Sun FUBARed the GUI, 
graphics, and security APIs to such an incredible extent that I don't 
expect them to ever recover.  For many cases, the improved APIs 
available to JavaScript produces comparable or better performance.

-- 
I will not see posts from astraweb, theremailer, dizum, or google
because they host Usenet flooders.

[toc] | [prev] | [next] | [standalone]


#29987

FromJosip Almasi <joe@vrspace.org>
Date2014-09-15 17:25 +0200
Message-ID<lv70cm$98f$1@gregory.bnet.hr>
In reply to#29985
On 09/15/2014 06:23 AM, Kevin McMurtrie wrote:
...
>
> I'd give up on running Java in a browser.

I did.

> Sun FUBARed the GUI,
> graphics, and security APIs to such an incredible extent that I don't
> expect them to ever recover.  For many cases, the improved APIs
> available to JavaScript produces comparable or better performance.

Yep; applet developpers, go GWT variants.
(I've gone to SmartGWT)

Regards...

[toc] | [prev] | [next] | [standalone]


#29991

FromKnute Johnson <eternal@knutejohnson.com>
Date2014-09-15 14:13 -0700
Message-ID<lv7kor$e9b$1@dont-email.me>
In reply to#29985
On 9/14/2014 21:23, Kevin McMurtrie wrote:
> In article <lv5fvs$12r$1@dont-email.me>,
>   Knute Johnson <nospam@rabbitbrush.frazmtn.com> wrote:
>
>> Two days of grief, my manifest file with only one line in it:
>>
>> Permissions:         sandbox
>>
>> Java kept rejecting the app as insecure.
>>
>> knute...
>
>  From what I've read in the Oracle bug database, Applet permissions
> essentially don't work any more.  I found this out by needing to get an
> Applet running - only one Java version works despite the Applet being
> correctly built and deployed.
>
> I'd give up on running Java in a browser.  Sun FUBARed the GUI,
> graphics, and security APIs to such an incredible extent that I don't
> expect them to ever recover.  For many cases, the improved APIs
> available to JavaScript produces comparable or better performance.
>

I love your website!  Wish I had found that a couple of weeks ago.  Had 
to find it all the hard way.

Thanks,

-- 

Knute Johnson

[toc] | [prev] | [standalone]


Back to top | Article view | comp.lang.java.programmer


csiph-web