Chrome bug with Applet and threading

R

Richard Maher

Hi,

Please see Chrome bug report: -
http://code.google.com/p/chromium/issues/detail?id=129528

If you can identify a bug in my code that is causing the hang on Chrome them
please let me know. Otherwise please vote for this issue by ticking the
star. SSCCE provided and behaviour very easy to reproduce.

Cheers Richard Maher

PS. If you can get it to work or fail with other OS/Browser combinations not
mentioned in the report (forget Safari on Windows) then please let me know.
 
R

Richard Maher

Evertjan. said:
Richard Maher wrote on 25 mei 2012 in comp.lang.javascript:

Off topic, Java != Javascript.

Can you please explain why you are so adamant that all of the Javascript on
the readily available: -
http://code.google.com/p/chromium/i...ken=uG9gNfEjUC8sTqGvGvrDTjpYD1E:1338292470931
is not responsible for the hang with the Chrome browser?

Perhaps the "dispatcher" method gets an exception and prevents the Applet
callback from calling notifyAll()? But that would surely only block one TAB?

Regardless, my decision to alter the Javascript so that the user's
"callback" JS function no longer has to call a "rendezvous" method but
rather return a true/false indicating end-of-stream for a reply/result-set
culminated in a solution that now works with every browser except Chrome.
Maybe a further tweak to the Javascript/DOM could get Chrome to obey the
rules? (Certainly appending the DIV child to the document.body *before*
seting its <object> innerHTML got IE8 over the line with Applet activation!)

Anyway, underneath all this is a single Java thread multiplexing output to
any number of TABs in a browser instance that constitute an Application, and
I think a lot of people would find that very interesting! Having said that,
to date, nobody could be bothered to tick a simple star so maybe no one
cares?

Oh, I get it! It's just that you can't wait till August when you and your
Cloggie mates get to take your caravans down to the autobahns and really
give complete strangers the shits?

Regards Richard Maher
 
S

Scott Sauyet

Richard said:
Can you please explain why you are so adamant that all of the Javascript on
the readily available: -http://code.google.com/p/chromium/issues/attachmentText?id=129528&aid...
is not responsible for the hang with the Chrome browser?

I did the same thing that Evertjan probably did. I looked at your
note, clicked on the link, scanned the bug report, saw absolutely
nothing in any of that sounded like a Javascript issue, getting turned
off in the process by the detailed descriptions of how to compile and
jar up your Java. (I don't know how the attachments work there, but
would it not be possible to attach the jar file as well as the
source?) I used to do mostly Java; I'm not scared of working with
Java or the command line, but that seems to me far from my current day-
to-day development with Javascript.

But I had no real idea what your code was supposed to do until your
follow-up here.

[ ... ]
Anyway, underneath all this is a single Java thread multiplexing output to
any number of TABs in a browser instance that constitute an Application, and
I think a lot of people would find that very interesting!

That could be interesting, I imagine. But it's been a long time since
I worked with an application that used multiple tabs.

Having said that, to date, nobody could be bothered to tick a simple star
so maybe no one cares?

I'm not likely to tick a simple star unless I've actually tried it and
saw the issue myself and decided that I thought it a reasonably high
priority. It's been eight years since I did *anything* with an
applet, so I really doubt that it would make it to that level.


[ ... ]
Oh, I get it! It's just that you can't wait till August when you and your
Cloggie mates get to take your caravans down to the autobahns and really
give complete strangers the shits?

Even though I have absolutely no idea what you're talking about here,
the obvious attitude makes me unwilling to go back and investigate
further.

-- Scott
 
L

Lew

Can you please explain why you are so adamant that all of the Javascript on
the readily available: -
http://code.google.com/p/chromium/i...ken=uG9gNfEjUC8sTqGvGvrDTjpYD1E:1338292470931
is not responsible for the hang with the Chrome browser?

You are asking the respondent why they are "so adamant" about a claim they've
never made. That is a logical fallacy and rhetorical dirty trick known as the
"straw man" argument.
Perhaps the "dispatcher" method gets an exception and prevents the Applet
callback from calling notifyAll()? But that would surely only block one TAB?

Regardless, my decision to alter the Javascript so that the user's
"callback" JS function no longer has to call a "rendezvous" method but
rather return a true/false indicating end-of-stream for a reply/result-set
culminated in a solution that now works with every browser except Chrome.
Maybe a further tweak to the Javascript/DOM could get Chrome to obey the
rules? (Certainly appending the DIV child to the document.body *before*
seting its<object> innerHTML got IE8 over the line with Applet activation!)

Anyway, underneath all this is a single Java thread multiplexing output to
any number of TABs in a browser instance that constitute an Application, and
I think a lot of people would find that very interesting! Having said that,
to date, nobody could be bothered to tick a simple star so maybe no one
cares?

What? What star? What are you talking about? What does that have to do with
Java or Javascript? That comment seems so random and non-sequiturish. And why
are you so aggrieved about whatever a "simple star" is?

Oh, I get it! It's just that you can't wait till August when you and your
Cloggie mates get to take your caravans down to the autobahns and really
give complete strangers the shits?

And thus you descend into simple trolldom. How is that a helpful remark,
"Richard"?
 
R

Richard Maher

Scott Sauyet said:
I did the same thing that Evertjan probably did. I looked at your
note, clicked on the link, scanned the bug report, saw absolutely
nothing in any of that sounded like a Javascript issue, getting turned
off in the process by the detailed descriptions of how to compile and
jar up your Java. (I don't know how the attachments work there, but
would it not be possible to attach the jar file as well as the
source?) I used to do mostly Java; I'm not scared of working with
Java or the command line, but that seems to me far from my current day-
to-day development with Javascript.

For those who don't want to dowload a JDK or worry about compiling anything,
there is a version of "POC.JAR" now attached to the Chrome bug report: -
http://code.google.com/p/chromium/issues/detail?id=129528#c9

Simply copy that file and the "hydra.html" file to your web hosting
directory and you're away. (To open the hydra.html example on other than the
local host you must first edit hydra.html and change the "codebase" and
"archive" object parameters from http://127.0.0.1/ to the correct URL for
your server.)

Cheers Richard Maher
 

Ask a Question

Want to reply to this thread or ask your own question?

You'll need to choose a username for the site, which only take a couple of moments. After that, you can post your question and our members will help you out.

Ask a Question

Members online

No members online now.

Forum statistics

Threads
473,743
Messages
2,569,478
Members
44,899
Latest member
RodneyMcAu

Latest Threads

Top